RESEARCH PAPER

Normal Values of Liver and Spleen Size by Ultrasonography

in Indian Children
BHAVNA DHINGRA, SUVASINI SHARMA, DEVENDRA MISHRA, #REEMA KUMARI, *RAVINDRA MOHAN PANDEY AND
#SHAILENDRA AGGARWAL

#
From the Department of Pediatrics and Department of Pediatric Radiology, Chacha Nehru Bal Chikitsalaya (Maulana Azad
Medical College),Geeta Colony, Delhi; and, *Department of Biostatistics, All India Institute of Medical Sciences, Ansari Nagar,
Delhi, India.
Correspondence to: Dr Devendra Mishra, Assistant Professor, Department of Pediatrics, Maulana Azad Medical College,
New Delhi 110 002, India. drdmishra@ gmail.com
Received: December 1, 2008; Initial review: January 12, 2009; Accepted: June 3, 2009.

Objectives: To establish normative data for the the participants.

ultrasonographic measurement of liver and spleen size in Results: Normal liver and spleen length and range were
healthy Indian children. obtained sonographically. The liver and spleen length
Setting: Tertiary-care pediatric teaching hospital. significantly correlated highly with the height/length of the
Design: Cross-sectional, hospital-based study. subjects (P=0.0001).

Participants: 597 healthy children between the ages of 1 Conclusions: The study provides the normal values of
month to 12 years. liver and spleen size by ultrasonography in healthy Indian
children.
Measurements: Ultrasonographic evaluation for the
assessment of liver and spleen size. These were Keywords: Children, Liver, Measurements, Organometry,
correlated with the age, sex, height/length and weight of Spleen, Ultrasonography.

Published online: 2009 September. PII:S097475590800697-1

L
iver and spleen size vary widely according establish standards of liver and spleen length by
to age. Many diseases can affect their size, ultrsonography in healthy Indian children, based on
ranging from infective processes to age, sex and somatometric parameters, as the
malignant disorders(1,2). Palpation and available literature is scarce(6,7).
percussion are the standard bedside techniques to
document liver and spleen size, but are far from Accompanying Editorial: Pages 475-476
accurate to detect small increase in size(1,2). The
spleen has to be enlarged two to three times its METHODS
normal size to be clinically palpable(3), although it
may be normally palpable in 15-17% of healthy We enrolled 650 asymptomatic children between the
neonates(4) and 10% of healthy children(3). Clinical age-group 1 month to 12 years, visiting the out-
assessment of hepatomegaly by palpation and patient department either for routine immunization
percussion has also been shown to lack both or accompanying their siblings, between January to
accuracy and reliability(2). December 2005. Any child under evaluation for/
follow-up case of a condition which could affect the
Ultrasonography is a non-invasive, established, size of the spleen or liver e.g. viral hepatitis, malaria,
safe, quick and accurate method for measurement of hemolytic anemia, enteric fever, congestive heart
liver and spleen size(5). We conducted this study to failure and malnutrition (Grade III and IV, IAP

INDIAN PEDIATRICS 487 VOLUME 47__JUNE 17, 2010

DHINGRA, et al. LIVER AND SPLEEN SIZE BY ULTRASONOGRAPHY

classification) was not included in the study. It was measurement of the liver span were marked and then
ensured by detailed structured parental history, measured from the sonographic image. The
examination and medical record review (if available) measurements were made during quiet breathing in
that these children did not have any pre-existing younger children and during breath-holding in older
suspected inflammatory, metabolic, traumatic, children. Neither preparation nor sedation was used.
collagen or hematopoietic diseases and
malignancies, that could affect liver and spleen size Statistical analysis: Differences of continuous
(excluded, n=11). Informed, written consent was variables between two independent groups were
obtained from the accompanying caregivers/parents assessed with the t test and the non-parametric
of all children, and verbal assent taken from all Mann–Whitney test; more than two groups were
children older than 5 year. No child refused assent, assessed with analysis of variance. Association
whereas there were 42 parental refusals for between spleen or liver length and each of the three
participation (in the majority due to lack of time to variables- age, height/length and weight was
spare, 85.7%). assessed with the Pearson correlation coefficient; to
identify the exact pattern of the relationship, non-
Baseline data including the age, sex, height/ linear regression analysis was performed. Multiple
length and weight were recorded for all the children regression analysis was applied in a backward
in a structured performa. The age was recorded to the stepwise fashion to test the independent effect of all
nearest completed month. An electronic weighing the factors on liver/spleen length. All statistical
scale (accuracy 50 g) and a wall mounted analyses were performed by the biostatistician using
stadiometer (1 mm markings) were used to measure STATA 10 software.
the weight and height/length, as per standard
methodology(8). In each child, the clinicians (SS) RESULTS
identified the acromioclavicular and the sternoclavi-
Parents of a total of 650 children were approached
cular joints by palpation and marked the midpoint of
for enrolment in the study, of which 42 refused
a line joining these two points– the midclavicular
permission and 11 were excluded. Thus, 597
point. A vertical line was drawn from the
children (347 [58.1%] boys) between the age-group
midclavicular point to the midinguinal point and was
1 month to 12 years (mean [SD] age 56.5 [41.9]
defined as the midclavicular line (MCL). All clinical
month) visiting the out-patient department either for
as well as sonographic measurements were done
routine immunization (268, 44.9%) or asymptomatic
with reference to this line(9).
children accompanying their siblings (329, 55.1%)
Ultrasonographic evaluation: All the children
underwent an ultrasonographic assessment of the TABLE I AGE AND SEX DISTRIBUTION OF THE STUDY
liver and spleen size on the same day by a single POPULATION
radiologist (RK), who was unaware of the clinical
Age group Male Female Total
condition and the observations made by the clinician,
using a Philips Envisor® Color Doppler system with 1-<3 mo 10 11 21
a multi-frequency 3.5 to 5 MHz probe. The 3-<6 mo 26 09 35
measurement of spleen length was the optically 6-12 mo 28 23 51
maximal distance at the hilum on the longitudinal
1-<2 yr 44 33 77
coronal view (between the most supero-medial and
the most infero-lateral points) as the spleen length at 2-<4 yr 79 53 132
the hilum is considered the most reproducible linear 4-<6 yr 62 53 115
measurement(10). The liver length was measured 6-8 yr 31 20 51
with the child in supine position and the section level 8-10 y 35 27 62
along the MCL was determined by simultaneous 10-<12 y 32 21 53
demonstration of the right kidney as per standard
methodology(11). The upper and lower points of the Total (%) 347 (58.1) 250 (41.9) 597

INDIAN PEDIATRICS 488 VOLUME 47__JUNE 17, 2010

DHINGRA, et al. LIVER AND SPLEEN SIZE BY ULTRASONOGRAPHY

were evaluated during the study. The median age was spleen size and the liver size increased significantly
48 month (range, 1-156 month). The age and sex with the age (P<0.05). Liver and spleen length
distribution of the study population is shown in correlated significantly with the height (r=0.84 and
Table I. The mean (SD) splenic length was 6.99 0.73) and weight (r=0.79 and 0.69). The regression
(1.36) cm (males, 7.06cm; females, 6.88cm) and the analysis with height/length as the independent
mean (SD) liver length was 9.59 (1.98) cm (males, continuous variable yielded a high multiple
9.63 cm; females, 9.54 cm). The spleen and liver correlation between spleen length and height/length
length of healthy children from 1 month to 12 year (R2=0.54), and liver length and height/length
according to age and sex is given in Table II. The (R2=0.70). Multiple regression analysis with spleen

TABLE II LIVER AND SPLEEN LENGTH BY ULTRASONOGRAPHY IN HEALTHY CHILDREN (N=597)#

Age and Sex No. Liver length (cm) Spleen length (cm)
Mean (SD) 3rd 97th Mean (SD) 3rd 97th
centile centile centile centile

1-<3 mo
M 10 6.5 (1.23) 4.8 8.9 4.9 (1.44) 3.7 8.7
F 11 6.2 (0.66) 4.9 7.2 4.5 (0.53) 3.2 5.2
3-<6mo
M 26 7.1 (0.77) 5.9 8.9 5.4 (0.61) 4.4 6.6
F 09 7.2 (0.94) 5.3 8.0 5.5 (0.51) 4.7 6.5
6-<12 mo
M 28 7.5 (0.88) 6.1 9.5 6.0 (0.86) 4.4 8.3
F 23 7.9 (0.92) 6.3 9.6 5.6 (0.61) 4.7 7.2
1-<2y
M 44 8.6 (0.85) 7.1 10.2 6.4 (1.01) 4.7 9.8
F 33 8.5 (1.51) 6.3 11.1 6.1 (0.74) 4.5 7.6
2-<4y
M 79 9.0 (1.34) 7.2 11.9 6.9 (1.01) 4.1 9.3
F 53 8.9 (0.97) 6.9 11.3 6.7(0.74) 5.3 8.3
4-<6y
M 62 10.3 (1.27) 7.3 14.7 7.4 (0.99) 5.0 10.9
F 53 9.8 (1.24) 6.5 13.3 7.1 (0.90) 5.2 9.3
6-<8y
M 31 10.8 (0.94) 09 12.3 7.9 (0.94) 6.3 9.7
F 20 10.9 (1.29) 8.2 13.3 7.6 (0.99) 5.5 9.5
8-<10y
M 35 11.9 (1.08) 10 14.1 8.2 (1.02) 6.8 10.9
F 27 11.7 (1.11) 9.4 14.0 8.2 (1.02) 6.5 9.8
10-<12y
M 32 12.6 (1.20) 11 15.5 8.7 (1.84) 6.3 11.7
F 21 12.3 (1.39) 9.7 15.2 8.7 (1.20) 6.7 11.3
# M-MALES, F-FEMALE, No difference between sexes either in the spleen size (P=0.11) or the liver size (P=0.57).

INDIAN PEDIATRICS 489 VOLUME 47__JUNE 17, 2010

DHINGRA, et al. LIVER AND SPLEEN SIZE BY ULTRASONOGRAPHY

length as dependent variable showed that both height across all ages and weights, in both the sexes.
(β coefficient 0.038, SE 0.006, P=0.0001) and body
surface area (β coefficient 3.776, SE 1.68, P=0.02) The conventional method of recording the
had significant independent positive association hepatic and splenic size by clinical examination has
with spleen length; no significant association could been reported to lack both accuracy and
be seen with age (β coefficient 0.0003, SE 0.003, reliability(1,2,7,12-14). There have been quite a few
P=0.91) or bodyweight (β coefficient 0.006, SE previous reports giving the standard sizes of liver
0.012, P=0.62). Multiple regression analysis with and spleen by ultrasound in children(5,6,9,11,15,16),
liver length as dependent variable showed that only but none has been done in Indian population. Among
height (β coefficient 0.055, SE 0.007, P=0.0001) had the more recent ones, Rosenberg, et al.(9) and
significant independent positive association with Megremis, et al.(5) have provided ultrasonographic
liver length; no significant association could be seen data on spleen size in 230 and 512 children,
with age (β coefficient 0.003, SE 0.004, P=0.39), respectively. Of the studies providing normative data
bodyweight (β coefficient 0.214, SE 0.014, P=0.14) on both liver and spleen size, Safak, et al.(16) studied
or body surface area (β coefficient –1.593, SE 1.97, 712, 7-15 year old children and provided data
P=0.42). Thus, only height had significant according to bodyweight groups. Konus, et al.(15)
independent positive association with both spleen studied liver and spleen sizes for 307 children and
length and liver length. The standardized ranges for provided the data in height-range and age-group
ultrasound organometry of liver and spleen in 1 categories. Our results are comparable to these
month-12 year old children, adjusted for height/ studies.
length and weight is provided in Table III. As the midclavicular line is known to vary widely
DISCUSSION when evaluated by different observers, we pre-
determined it for the sonographic assessments, thus,
Our results provide a standard set of normal range of consistency of measurement was ensured. A single
liver and spleen size according to weight, height, age radiologist performing the ultrasonograhy removed
and sex of the children, as determined by the inter-observer bias. There are also some
ultrasonography. We also found height to be a limitations of our study. The number of subjects in
significant correlate of the liver and spleen size the age group 1-3 months and the number of females

TABLE III LIVER AND SPLEEN SIZE ADJUSTED FOR HEIGHT/LENGTH AND WEIGHT ON ULTRASOUND ORGANOMETRY

Parameter No Age (mo) Sex ratio Liver size (cm) Spleen size (cm)
mean (SD) (female/male) mean (SD) mean (SD)

Height/length (cm)
50-80 151 10.5 (6.7) 90/61 7.6 (1.18) 5.7 (0.99)
81-100 174 38.4 (11.6) 98/76 8.9 (1.14) 6.7 (0.90)
101-120 151 71.9 (17.5) 90/61 10.3 (1.19) 7.4 (0.99)
121-140 87 114.7 (16.5) 46/41 11.9 (1.18) 8.3 (1.0)
>140 34 141 (8.6) 23/11 12.5 (1.32) 8.8 (1.11)
Bodyweight (kg)
≤10 143 10.5 (7.0) 59/84 7.5 (1.12) 5.6 (0.90)
10.1-20 310 50.1 (21.1) 132/178 9.5 (1.38) 7.0 (1.03)
20.1-30 100 107.1 (20.3) 44/56 11.5 (1.19) 8.1 (0.99)
30.1-40 33 134.5 (12.3) 11/22 12.4 (0.92) 8.6 (0.99)
>40 11 141.8 (12.9) 4/7 13.7 (1.02) 9.4 (1.08)

INDIAN PEDIATRICS 490 VOLUME 47__JUNE 17, 2010

DHINGRA, et al. LIVER AND SPLEEN SIZE BY ULTRASONOGRAPHY

WHAT IS ALREADY KNOWN?

• Ultrasound-based organometric data is available for different population groups.

WHAT THIS STUDY ADDS?

• Ultrasound-based organometric normative data on the liver and spleen size in Indian children aged 1 month
to 12 year is presented, and shown to correlate with the height/length of the child.

in the age-group 3-6 months was less than ten. Contributors: SS and DM conceptualized the study. BD,
Further, neonates, and children above 12 year age SS and DM were involved in patient selection, enrolment
were not included in the study population due to and clinical data collection. RK and SA conducted the
administrative constraints. Although, many previous sonographic evaluation. SS and BD collated the results,
conducted the literature search and drafted the manuscript.
studies have published normative data using similar
DM supervised the conduct of the study and the
small numbers(5,9), the small sample size in certain manuscript preparation. RMP was involved in the
groups may affect the generalizability of the values planning of the study, conducted the statistical analysis,
to these age-groups. Nutritional anemia and and reviewed the manuscript for important intellectual
malnutrition may be associated with organomegaly content. All the authors were involved in the approval of
and were not specifically excluded (except severe the final manuscript to be published. RK and DM would
malnutritiona and clinically obvious anemia). act as the guarantors for the relevant portions of the
However, by presenting the largest pediatric series manuscript.
so far, we provide a more accurate assessment of Funding: None.
liver and spleen sizes in children, especially those in Competing interests: None stated.
India.
REFERENCES
In addition to size, there are several palpatory
1. Zhang B, Lewis SM. A study of the reliability of
characteristics of the liver and spleen (tenderness, clinical palpation of the spleen. Clin Lab Haematol
liver edge, nodularity and consistency of the surface, 1989; 11: 7-10.
etc.) that contribute significantly to the overall
2. Joshi R, Singh A, Jajoo N, Pai M, Kalantri SP.
bedside assessment of the organomegaly. Thus,
Accuracy and reliability of palpation and
clinical liver span remains a simple practical percussion for detecting hepatomegaly: a rural
measurement of liver size, also providing additional hospital based study. Indian J Gastroenterol 2004;
supplemental information, and the most applicable 23: 171-174.
in developing countries. The bedside assessment of
3. French J, Camitta BM. Splenomegaly. In: Behrman
liver and splenic enlargement will not obviate RE, Kliegman RM, Jenson HB (ed). Nelson
diagnostic imaging when such information is vital to Textbook of Pediatrics. 17th ed. Philadelphia, Pa:
further therapeutic management of the patient. Saunders; 2004. p.1675.

We believe that the results of this study can be 4. Mimouni F, Merlob P, Ashkenazi S, Litmanovitz I,
used as a practical and comprehensive guide to Reisner SH. Palpable spleens in newborn
term infants. Clin Pediatr (Phila) 1985; 24: 197-
indicate the normal liver and spleen length range for
198.
every child, according to his/ her age and body
habitus. 5. Megremis SD, Vlachonikolis LG, Tsilimigaki AM.
Spleen length in childhood with US: Normal values
ACKNOWLEDGMENTS based on age, sex and somatometric parameters.
Radiology 2004; 23:129-134.
Dr KK Kalra, Medical Superintendent, Chacha 6. Chen CM, Wang JJ. Clinical and sonographic
Nehru Bal Chikitsalaya, for the permission and assessment of liver size in normal Chinese
logistic support to conduct and publish the study. neonates. Acta Paediatr 1993; 82:345-347.

INDIAN PEDIATRICS 491 VOLUME 47__JUNE 17, 2010

DHINGRA, et al. LIVER AND SPLEEN SIZE BY ULTRASONOGRAPHY

7. Singh K, Bhasin DK, Reddy DN, Koshy A. Liver 13. Halpern S, Coel M, Ashburn W, Alazraki N,
span in normal Indians. Indian J Gastroenterol Littenberg R, Hurwitz S, et al. Correlation of liver
1985; 4: 73-75. and spleen size. Determinations by nuclear
8. Swash M. Hutchison’s Clinical Methods. 20th medicine studies and physical examination. Arch
edition. London: WB Saunders Company Ltd; Intern Med 1974; 134: 123-124.
1995. 14. Barkun AN, Camus M, Green L, Meagher T,
9. Rosenberg HK, Markowitz RJ, Koelberg H, Park Coupal L, De Stempel J, et al. The bedside
C, Hubbard A, Bellah RD. Normal splenic size in assessment of splenic enlargement. Am J Med
infants and children: sonographic measurements. 1991; 91: 512-518.
Am J Roentgenol 1991; 157: 119-121.
15. KonuÛs OL, Ozdemir A, Akkaya A, ErbaÛs G, Celik
10. Deligeorgis D, Yannakos D, Panayoton P. Doxiadis H, IÛsik S. Normal liver, spleen, and kidney
S. The normal borders of the liver in infancy and dimensions in neonates, infants, and children:
childhood. Arch Dis Child 1970; 45: 702-704. evaluation with sonography. AJR Am J Roentgenol
11. Dittrich M, Milde S, Dinkel E, Baumann W, 1998; 171: 1693-1698.
Weitsel D. Sonographic biometry of liver and 16. Safak AA, Simsek E, Bahcebasi T. Sonographic
spleen size in childhood. Pediatr Radiol 1983; 13: assessment of the normal limits and percentile
206-211. curves of liver, spleen, and kidney dimensions in
12. Kirk RM. Clinical assessment of liver enlargement. healthy school-aged children. J Ultrasound Med
Indian J Gastroenterol 2004; 23: 163-164. 2005; 24: 1359-1364.

INDIAN PEDIATRICS 492 VOLUME 47__JUNE 17, 2010