Abstract
Background
Knowledge of the normal sizes of the liver, spleen, and kidneys is important to radiologists when assessing for pathology using ultrasound scan. The need for a local determination of a easy-to-use formula for estimating the expected normal sonographic dimensions of these organs in children in order to serve as baseline when assessing them for pathology cannot be over emphasized.
Purpose
To determine ultrasonographic sizes of the liver, spleen, and kidneys among primary school children in southwestern Nigeria and correlate these with anthropometric measures to provide local reference data and an easy-to-use formula for assessing these organs for pathology in clinical practice.
Material and Methods
This is an observational, cross-sectional study using 1000 public primary school children in Ogbomoso. Sonographic dimensions of their liver, spleen, and both kidneys with anthropometric parameters were obtained and correlated. Data were analyzed using SPSS version 20.
Results
The liver span was higher in boys than in girls (P = 0.048) while the left lobe of the liver was higher in girls than in boys (P = 0.003). The spleen length was higher in boys than in girls (P = 0.011). There was no gender difference in kidney dimensions (P > 0.05). All anthropometric measures correlated significantly (P < 0.001) with organ dimensions. Body surface area is the strongest predictor of the liver and kidney sizes (P < 0.001) and height for spleen size (P < 0.001).
Conclusion
Nomograms for the liver, spleen, and kidneys and regression equations for estimating the dimensions of these organs were formulated based on the best multivariate correlates.
Introduction
Knowledge of the normal dimensions of the liver, spleen, and kidneys is not only important to radiologists but also to clinicians when assessing these organs for pathology. This is especially important in the pediatric age group due to the organ size increasing with age. The dimensions of organs have been shown to change with pathologic states: while some diseases cause an increase in size, others cause a reduction in size and the direction of change in organ size can be used to predict the prognosis of some diseases (1–3).
Many studies have shown that the size and/or measurements of the liver, spleen and kidneys are influenced by many factors such as age, ethnicity, race, gender, weight, and height (4,5). It has been estimated that between birth and adulthood, there is an increase in liver mass of least tenfold (6,7). Hepatomegaly is a frequent clinical finding in children and may be caused by intrinsic liver diseases or by systemic alterations. In case of clinical suspicion, ultrasonography (US) is generally the first method of choice when investigating pediatric patients (8). In a similar vein, altered spleen size can be due to infective, infestation, infiltrative, immunologic, and malignant conditions; examples of these diseases are malaria, tuberculosis, typhoid fever, and other water-borne parasitic infections that are a major public health concern in Nigeria (2,4,9). In 2003, it was estimated that half of the Nigerian population has at least one episode of malaria annually and the majority of outpatient visits can be attributed to malaria (3). Evaluation of kidney measurements such as length, width, and thickness is an important parameter in the diagnosis as well as management of many kidney disorders as it is known that there is a close relationship between the size of the kidney and its function (10,11).
In common practice, measurements of the size of these organs at any age are compared with the dimensions that are predicted by standard nomograms. The current normal graphs that are widely used locally are derived largely from studies based on Caucasian populations of relatively small sample sizes (5,8).
Therefore, the aims of the present study were to bridge the gap in knowledge by formulating an age–specific, anthropometric-correlated nomogram of liver, spleen, and kidney measurements and to develop a regression equation that can be easily used by both the sonologist and the physician, by the bedside, when assessing these organs for pathology and monitoring response to therapy among a Nigerian pediatric population in the southwest.
Material and Methods
This is a descriptive, cross-sectional study of liver, spleen, and kidney parameters among Nigerian children in primary schools. It was carried out at public primary schools in Ogbomoso metropolis from March 2016 to October 2016.
A multi-staged sampling method was used. A total of 1000 pupils (500 boys, 500 girls; age range = 5–13 years) were recruited into the study. Based on proportionate sampling, 500 children were recruited from each of the two local governments in Ogbomoso. Five primary schools were chosen in each local government by simple random sampling. Pupils were selected by systematic random sampling among volunteers from each class (primary 1–6) based on the school’s population and number of children in each school.
Exclusion criteria
The clinical exclusion criteria included: refusal to participate; jaundice; fever (T > 37.5 °C); maculopapular rash; significant lymphadenopathy; short stature (<2 SD for age); limb deformity; malnutrition; obesity; pallor; macrocephaly (head circumference >2SD for age); microcephaly (head circumference <2SD for age); chest wall deformity; and known liver, spleen, kidney, and systemic diseases.
The imaging exclusion criteria included: liver, spleen, and kidney parenchymal mass lesions; abnormal parenchymal echotexture and echogenicity; liver, spleen, and kidney cysts; accessory spleens; and hydronephrosis.
Data collection instrument
A semi-structured questionnaire was used. This was pretested among 10 children attending a public primary school outside of the selected schools in Ogbomoso and it was administered by the interviewer. Data collection also used tape measures, weighing scales, examination couches, and acoustic gel.
Demographic data were collected from each participant at the time of their pre-participation physical examination. This information included: age; sex; height and body weight; body surface area (BSA); and body mass index (BMI), computed from measured height and weight. BMI was calculated as weight (kg)/height (m2) while BSA was calculated as √weight ×height/3600. Anthropometric measurements were obtained on the participants wearing school uniform without shoes. Weight was measured on a calibrated portable Salter scale to the nearest 0.1 kg. Height was measured with a metal tape measure to the nearest 0.5 cm with the participants standing upright with the head in the Frankfurt position.
Scanning was done with the individual lying in the supine position, while the area scanned was exposed. The participant’s right upper limb was then raised behind the head to help widen the intercostal spaces and the space between the lower costal margin and the iliac crest, thereby creating better access to the liver (12). Scanning was done in both sagittal and transverse views by two radiologists.
Using a 2.5–5 MHZ curvilinear transducer of a full digital portable ultrasound diagnostic device (Sonoace X1, Shangai Medison Medical instruments Co. Ltd, China), the diagonal length of the liver (liver span) was measured in the supine position while the length of the spleen was measured in the right lateral decubitus position. The length, width, and thickness of the kidneys were measured in the prone position. The dimensions of these organs were taken during deep inspiration. All measurements were obtained to the nearest millimeter on static original ultrasound images using electronic calipers at the time of scanning. Two sequential measurements were obtained for each organ dimension with the image frozen in inspiration, and their mean was calculated; this was to minimize intra-observer error and ensure greater accuracy and reliability of the measurements (12,13).
According to the methods described earlier (6–9), the diagonal axis of the liver from its most inferior aspect on the right to the most lateral aspect on the left was measured as the liver span (Fig. 1). The technique of the right lateral decubitus position in the coronal plane was adopted for the spleen length measurement. Longitudinal size measurement was performed between the most superomedial and the most inferolateral points of the spleen (Fig. 2). The maximum length of each kidney was measured between the uppermost edge of the upper pole and the lowest edge of the lower pole (bipolar diameter) of the kidney and the anteroposterior (AP) diameter was measured between the most anterior and posterior capsule while the participant lies in the prone position. The probe was then rotated 90° and cross-sectional measurements of the kidney width at the hilar level was performed (Figs. 3 and 4). These measurements were obtained with the patient lying prone. Kidney volumes were calculated by using the standard formula for ellipsoid structures (volume = length × width × AP diameter × 0.52).
A B-mode longitudinal ultrasound image through the right lobe of the liver at the mid-clavicular line showing the liver and the right kidney. A to B = liver span.
A B-mode coronal ultrasound image through the spleen showing the spleen and the left kidney. A to B = splenic length.
A B-mode longitudinal ultrasound image through the right kidney showing the kidney and part of the right lobe of the liver. A to B = kidney length, C to D = anteroposterior dimension of the kidney.
A B-mode transverse ultrasound image through the right kidney showing the kidney and part of the right lobe of the liver. A to B = kidney width.
The data sheet was sorted out manually. Data from the questionnaires and the liver, spleen, and kidney ultrasonographic measurements were entered into statistical package for Social Scientists (SPSS) computer software, version 20 (SPSS, Chicago, IL, USA). Frequency distribution tables were used to present results. The chi-square test was used to test association between qualitative variables. The Student’s t-test was used as a test of association between two continuous variables while analysis of variance (ANOVA) was used when there were > 2 continuous variables. Level of significance was set at P < 0.05.
Ethical consideration
Ethical clearance was obtained from the local review board of the hospital. Permission was taken from the respective school board. Written informed consent was sought from the parent of each child and any parent who declined had his/her child excused from the study. The parents were reassured that no harm would be done to their children.
Results
Of the study population, 971 (97.1%) were of Yoruba ethnicity, 19 (1.9%) were Hausa, 6 (0.6%) were Igbo, and 4 (0.4%) were from other tribes.
The mean age of the girls was 9.37 ± 2.3 years while that of the boys was 9.36 ± 2.3 years (P = 0.978). The average height of the participants was 128.6 ± 12.8 cm (range = 99.0–173.0 cm) and the average weight was 26.8 ± 7.2 kg (range = 16.0–55.0 kg). The calculated average BMI was 16.0 ± 2.1 kg/m2 and the average BSA was 0.97 ± 0.18 m2 (range = 0.52–1.60 kg/m2). There was no statistically significant difference between the age, height, weight, BMI, and BSA of both sexes (P = 0.978, 0.352, 0.720, 0.658, and 0.521, respectively); thus, boys and girls were anthropometrically matched.
The liver span, spleen length, kidney length, and volume by age, height, weight, BMI, and BSA are as depicted in Tables 1 and 2.
Distribution of spleen and liver dimensions by age and sex.
Distribution of prone right kidney length and volume by age and sex.
Table 3 shows the relationship between anthropometric parameters and liver span, spleen length, kidney length, and volume on multivariate analysis. BSA indicated significantly strong correlation of the univariate analysis with liver span, kidney length, and kidney volumes (P < 0.001), while age (P = 0.016) and height (P < 0.001) indicated significantly strong correlations of the univariate analysis with splenic length.
Multivariate regression analysis showing the relationship between liver, spleen, and kidney dimensions with age, height, weight, BMI, or BSA.
*Standardized coefficients of regression.
BMI, body mass index (kg/m2); BSA, body surface area (m2).
Based on the evidence from the multivariate analysis, regression equations were developed for the bedside assessment of the liver span, spleen length, kidney length, and kidney volumes in children as depicted in Table 4.
Simple regression formula for the liver, spleen, and kidney (prone) dimensions using the body size indicators based on the best model.
BSA, body surface area; H, height.
Discussion
The evaluation of liver, spleen, and kidney measurements is very important to the clinician as the result can be used as baseline when assessing these organs for pathology. Earlier studies carried out on these organs in Nigerian children were mainly among those of Igbo ethnicity (age range = 5–17 years) (2,12). The present study was conducted among children who were mainly of Yoruba ethnicity (age range = 5–13 years) (97.1%).
Similar findings by Ezeofor et al. (12) and Farheen (14), that the liver dimensions were higher in boys, is observed in the present study, but is in contrast with other studies where no gender variation was found (2,4,6,8,9,15,16). An earlier study on the ultrasound dimensions of the liver among southeastern Nigerian children by Eze et al. (15) also quoted a higher liver dimension in boys. This contrast might be due to the difference in physique, which is one of the factors that determines liver dimensions in people.
On multivariate analysis, the liver span was found to correlate significantly with BSA and BMI. This is similar to the report from southeast Nigeria in which BSA correlated the most with the liver dimensions followed by weight (12). BSA and BMI have, however, been reported to correlate inconsistently with liver dimensions by some other researchers outside of Nigeria (17–19). Height and weight have also been found by many other researchers to correlate significantly with liver dimensions (18–20).
The splenic size in the present study did not consistently increase with age. A researcher reported a similar finding in the relationship of splenic length with age among Sudanese children (11). This is contrary to earlier reports from southeast Nigeria that showed an increase in splenic size with age in both sexes (12). Salam et al. (1) also reported an increase in splenic size with increasing age in a study carried out in Egypt. Age, therefore, may not be used in predicting splenic length among children from the southwestern part of Nigeria. This difference might be due to genetic factors.
The splenic length in the present study is lower than that reported among children from southeastern Nigeria (12), Sudan (11), and Pakistan (14). This is evidence of ethnic and racial differences. Ezeofor and other researchers reported significantly larger spleen dimensions in boys (5,12,22); this is in agreement with the finding in the present study. Nouri et al. (11), however, reported a larger splenic dimension in female Sudanese children compared with their male counterparts. However, some studies did not report any sexual dimorphism in splenic dimensions (17–21).
On multivariate analysis, height correlated best with splenic length in the present study followed by age; however, BSA was found to correlate best among children from the southeast of Nigeria (12). Salam et al. (1) is in agreement with the present study: they reported that height showed the best correlate with splenic length. In an earlier study among Nigerian adults from the southeast, Ehimwenma and Tagbo (22) reported weight as the best correlate. Physique and diet may be responsible for these differences.
It is well-known that kidney size is related to age, height, and weight. The present study also shows a significant relationship between kidney parameters and anthropometric measures. Some studies have shown that height correlates best with kidney length (23–27) while others revealed that weight is the best correlate with kidney length. In the present study, however, BSA correlates best with kidney length and volume, followed by weight and height.
It is believed that, after a wide search, no easy-to-use bedside formula for the assessment of these organs has been generated. Therefore, a regression equation for estimating the dimensions of these organs was formulated based on the best multivariate correlates for each organ.
The present study has some limitations. First, the study was carried out in a limited geographical location; hence, a multicenter international collaboration with larger sample size may be advised to further validate the strength of the regression formulae in children, especially those outside the study location. It is also worth noting that the study was carried out among children attending public schools. This may have skewed the study population towards children from homes with a low and middle socioeconomic status. A further study that is more inclusive will be advocated. In the same vein, blood and urine samples may have been collected to screen the children for subclinical abnormalities in the studied organs.
In conclusion, the nomograms generated from the present study and the formulated regression equation can be referred to when assessing these organs for pathology especially among children from the southwest of Nigeria. Further studies in other centers will be necessary in others to identify the factors responsible for the differences in organ dimensions of children in southeast and southwest Nigeria
Footnotes
Acknowledgements
The authors thank Akinlade Olawale Mathias for his assistance during the execution phase of the project and write-up.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
