This was a prospective longitudinal study. The subjects were women who gave birth at the University of Nigeria Teaching Hospital, Ituku Ozalla, Enugu State, the foremost tertiary health institution in Southeast Nigeria, between November 2012 and May 2013. Ethical approval (NHREC/05/01/20088B-FWA00002458-IRB00002323) for the proposed study was obtained from the Health Research Ethics Committee of the University of Nigeria Teaching Hospital, Ituku Ozalla, Enugu State before the commencement of the study.

Sample size calculation: The minimum sample size was calculated using Cochrane’s formula: n=Z² p (1-p)/d²

for a population greater than 10,000 at 95% confidence interval (z=1.96), 50% prevalence level (p), and absolute precision (d) of 5%. The calculated minimum sample size was approximated to 400 participants.

Inclusion and Exclusion criteria: All women who had uncomplicated singleton pregnancies that were carried to term with uneventful deliveries and puerperium were recruited for this study. Women with morbidities during pregnancy and delivery were excluded from the study. Women with pre-pregnancy pathologies (leiomyoma uteri, adenomyosis, etc.), scars from surgeries (Caesarean section, myomectomy, etc.), and those that did not complete the study were excluded from the study. If any abnormality was noted the patient was referred for the appropriate management and excluded from the study.

Before enlistment into the study, informed consent was sought and obtained from each subject and the procedure was properly explained. The age, parity, breastfeeding pattern, type of delivery and past medical history was obtained by direct questioning and proper history taking.

Transabdominal sonography was carried out on each subject using a portable ALOKA ultrasound machine {made in Japan by ALOKA Ltd., model IPC-1530(u); SN model 7292} with a 3.5-5. MHz curvilinear electronic transducer. The ultrasound examinations were done on a postpartum day 1, day 3 and day 42. Day 1 and day 3 were chosen to assess uterine involution in the early puerperium. Day 42 was chosen for assessment of uterine involution at late puerperium and also because postpartum women come for their routine six weeks visit on that day which reduces the number to be lost to follow-up. The subjects were made to lie in a comfortable supine position and with the bladder moderately filled with urine which served as an effective acoustic window especially for day 42 examination. This helped to visualize the uterus and endometrium better as bowel loops containing gas were displaced upwards away from the pelvis. The subjects were usually asked to drink about 2.5 liters of potable water to get the bladder moderately filled.

Coupling inert gel was used to eliminate air between the probe surface and the skin. Gentle compression with the probe was used and the measurements were taken between uterine contractions especially on day 1 and day 3. On day 42 there are usually no contractions so the measurements were taken without emphasis on uterine contractions. The uterus and uterine cavity were assessed in longitudinal and transverse sections. In the sagittal scan, longitudinal diameters (the length) and anteroposterior diameters (thickness) of the uterus and uterine cavity were measured with the anteroposterior diameters perpendicular to the length. The longitudinal diameter (length) of the uterus was taken from the highest fundal point in the midline to the external os of the cervix. The anteroposterior diameter was measured from the anterior to the posterior wall and perpendicular to the length. The transverse diameter (width) of the uterus was the widest diameter from right to left in the midline on a transverse scan [10,12]. The uterine cavity (endometrium) was critically assessed to rule out retained products of conception.

The analysis was carried out using the SPSS software version 20.0 for descriptive statistics, central tendencies and cross-tabulation result display. The uterine dimensions were expressed as mean±standard deviation and student’s t-test analysis was used to compare the values for day 1, day 3 and day 42 of puerperium. Regression analysis was carried out to measure the relationship between uterine measurements on days 1, 3 and 42. A p-value of <0.05 was considered statistically significant.

The mean age was 31.1±4.78 years. The highest proportion of the subjects, 156 (39%) were in the 30-34 year age category. Majority of them were multiparous 253 (63.2%), gave birth by vaginal delivery; 234 (58.5%) and practiced exclusive breastfeeding; 316 (79%) [Table/Fig-1].

The mean diameter of the uterus (longitudinal, anteroposterior and transverse) on day 1, day 3 and day 42 of puerperium is shown in [Table/Fig-2]. All diameters reduced with increasing days of puerperium.

There was a reduction in all three uterine diameters with increasing days of puerperium and were significant. The most rapid rate of reduction of all diameters occurred between days 1 and 3. There was a marked reduction in the rate of involution between days 1-3 and days 3-42 across all three diameters (7.54, 5.81 and 2.83 for Longitudinal, Anteroposterior and Transverse diameters, respectively) [Table/Fig-3].

The mean diameters of the uterine cavity (longitudinal, anteroposterior and transverse diameters) showed a reduction with increasing days of puerperium [Table/Fig-4].

The most rapid rate of reduction of all diameters occurred between days 1 and 3. There was a marked reduction in the rate of involution between days 1-3 and days 3-42 across all 3 diameters (7.54, 26.67 and 10.00 for Longitudinal, Anteroposterior and Transverse diameters, respectively) [Table/Fig-5].

As shown in [Table/Fig-6], measurements of all three dimensions (longitudinal, anteroposterior and transverse) on day 1 can be used to predict what these dimensions will be on day 3. In the same way, anteroposterior dimensions on day 1 and day 3 were predictive of day 42 dimensions.

Similarly, day 1 measurements of the uterine cavity (longitudinal, anteroposterior and transverse) can be used to predict day 3 measurements. Also, the longitudinal diameter on day 1 was found to be a predictor of the day 42 diameter.

The most rapid involution of the uterus occurs within the first week of puerperium [13]. Regeneration of the endometrium is complete by the third week postpartum but regeneration of placental site is not complete until 5-6 weeks [12]. In this present study, the mean age was 31.1±4.78 years and the predominant age group was 30-34 years. This is comparable to the study by Olayemi O et al., on 300 postpartum women in the southwestern part of Nigeria [12] and is in keeping with other studies in different parts of the world [7,14].

The mean uterine longitudinal diameter on day one was 14.5±1.27 cm. Although it was lower than in some other studies [12,15,16], it is still within the normal expected value for uterine length postpartum which ranges from 14.5 to 25 cm [17]. Olayemi O et al., in South West Nigeria [12] reported a value of 17.1 cm on day 1 while Middleton WD et al., and Sokol ER et al., reported 19.9 cm and 16.1 cm, respectively [15,16]. These were carried out in the United States of America and the differences noted in these various studies could be due to racial and anthropometric variations between the study populations. In this study, all diameters of the uterus reduced significantly as puerperium progressed. This is in keeping with what is expected in uterine involution; as puerperium progresses, the uterus contracts and reduces in size until the 6^th week when it returns to its prepregnant size. Several studies have shown that uterine diameters, regardless of the particular diameter studied, reduces as puerperium progresses [12,14-16,18,19]. Galli D et al., stated that the reduction in the diameters and volume of the uterus was more evident during the first four days [18]. As observed in this study, It has also been reported that the length and anteroposterior diameters reduce more quickly than the transverse diameter [1]. However, controversy still exists on which diameter is the best for assessing uterine involution [7,14,19].

Mulic-Lutvica A et al., reported that the anteroposterior diameter of the uterus diminished substantially and progressively from 92.0 mm on day 1 postpartum to 38.9 mm on day 56 [7]. Their findings were similar to the 93.0 mm (day 1 postpartum) and 38.6 mm (day 56) reported by Abdel-Nabi A et al., both authors also reported that the anteroposterior diameter of the uterine cavity showed a progressive reduction until day 56 [14]; reducing from 15.8 mm on day 1 to 4.0 mm on day 56. On the other hand, the anteroposterior diameter of the uterine cavity measured 5 cm from the fundus, however, showed an increase on days 7 and 14 postpartum. Uniquely, during the normal puerperium the position, shape and cavity of the uterus undergo some changes.

All diameters of the uterine cavity reduced significantly with increasing days of puerperium. This was an expected finding and could be explained by the fact that the process of the uterus contracting during involution also leads to gradual expulsion of the uterine cavity contents. Although the mean uterine cavity anteroposterior diameter of 0.77 cm on day 1 was smaller than a number of other studies [7,12,14]. It is still within normal diameter for postpartum uterine cavity involution which is less than 2 cm. The differences in these various studies could be due to the difference in the racial and anthropometric variables of the population studied.

Klug PW was of the opinion that an exact involution of the length of the uterus can be made only by sonography [20]. Middleton WD et al., reported that at mean times of 1.4 days, 2.7 weeks and 6.7 weeks after delivery, the uterus decreased in length from a mean of 19.9+2.1 cm to 11.2+2.0 cm and then to 8.7+1.2 cm, respectively [15]. They also noted that the endometrial cavity was usually empty as at first week of puerperium and also on day 28. However, haematometra is usually seen on day 7 and give rise to no clinical signs.

Belachew J et al., studied uterine involution with three dimensional ultrasound by measuring uterine body and cavity volumes [21]. They reported a marked reduction in the median uterine body volume from 756 cm³ on day 1 to 68 cm³ on day 56 and in the median cavity volume from 22 cm³ on day 1 to an insignificant volume on day 56.

In this present study, there was a significant relationship between day 1 and day 3 longitudinal and anteroposterior diameters. This implies that day 3 longitudinal and anteroposterior diameters can be predicted using day 1 diameter. As for day 42, the diameter cannot be accurately predicted because of the regression between day 1 and day 42 was not significant. The linear regression in the transverse diameter showed no significant relationship between day 1 and days 3 and 42. The diameters for day 3 and day 42 cannot be accurately predicted using day 1 diameter. In this study, the linear regression between longitudinal uterine cavity diameters showed significant relationship between day 1 and day 3 and day 1 and day 42. Conversely, the linear regression between anteroposterior and transverse uterine cavity diameters also showed no significant relationships for both days 3 and 42. This implies that ultrasonography in these women on day 3 can be predicted if day 1 parameters are stated (longitudinal) whereas the anteroposterior and transverse diameters cannot be accurately predicted and so must be measured and not estimated.

This study was an institution based; the participants selected for the study may not be true representation of the actual population of Southeastern Nigeria.

This study has established a normal range of the diameters of the uterus and the uterine cavity on days 1, 3 and 42 of puerperium and there is reduction in these diameters with increasing days of puerperium. The study has served as a baseline for future studies through the provision of reference data. A study to compare the use of 2 and 3-dimensional sonography for the evaluation of uterine involution in puerperium is also recommended to ascertain if there is any advantage of volumetric measurements over numeric measurements.