ORIGINAL

Niger J Paed 2013; 40 (1):55 –59

Animasahun BA

Njokanma OF

Nwodo UP

Relationship of age, anthropometry

and haemoglobin concentration with

echocardiographic findings in

Nigerian children with sickle cell

anaemia.

Omokhodion SI

DOI:http://dx.doi.org/10.4314/njp.v40i1,10

Accepted: 30th April 2012

Abstract Background: Assessment

of the structural and functional ef-

fects of sickle cell anaemia (SCA)

on the heart can be done using echo-

cardiography. There has been no

agreement on which indicator of

body size (weight, height, and body

surface area (BSA)) or age in chil-

dren should be used to relate to and

correct echocardiographic measure-

ments.

Aim: To correlate the echocardio-

graphic findings of children with

SCA aged one to fifteen years in

steady state with their age, anthro-

pometry and haemoglobin concen-

tration.

Methods: A prospective study car-

ried out at the Lagos University

Teaching Hospital (LUTH), be-

tween May and October 2005. The

subjects were 60 paediatric patients

attending the LUTH SCA outpatient

clinic and 60 age, sex and socio-

economic matched controls. Their

height, weight, and echocardio-

graphic parameters were measured

and their BSA and haemoglobin

level were determined.

BA Animasahun

Njokanma OF

(

)

Results: The direct echocardio-

graphic measurement (AO, LA,

LVPW, EDD, ESD, IVS) and LVM

each increased progressively with

body weight, height, body surface

area and age in both subjects and

controls (r > 0.3, p < 0.001 in each

case). Multivariate analysis showed

that height correlated significantly

with AO, ESD, EDD and LVPW in

the subjects. All the parameters

except age were significantly corre-

lated with LVM both in univariate

and multivariate analysis.

Conclusion: Echocardiographic

parameters in SCA patients are sig-

nificantly correlated with anthro-

pometric indices and age. Our data

suggest best correlation with height

in these subjects.

Department of Paediatrics and

Child Health

Lagos State University College of

Medicine

Ikeja. Lagos. Nigeria.

Tel: +2348037250264

Email: deoladebo@yahoo.

Nwodo UP

Premier Specialist Hospital

Victoria Island, Lagos

Phone number: 2348033002402

Email: rolinwodo@doctor.com

Omokhodion SI

Department of Paediatrics

University College Hospital

Ibadan, Nigeria.

Tel: 2348023151475

Email: samodion@yahoo.com

Keywords: Sickle cell anaemia,

Echocardiographic parameters, An-

thropometry, Steady state

.

Introduction

with age and weight.

Sickle cell anaemia affects practically every system in

the body and may sometimes present as crisis situations.

Some of the chronic manifestations involve the cardio-

vascular system₁_-₆ as a result of chronic anaemia and re-

In conformity with the previous studies, Sutton, Pickard,

Oldershaw et al also documented increased cardiac

dimensions with age in a group of 78 normal children

aged one to 12 years.

9

current crisis.

Different studies in normal children

1

0

have shown the fact that cardiac dimensions₇_-₈increase

Oladokun in her study of 183 normal children aged 1

to 12 years in Ibadan noted that all the direct echocar-

diographic measurements in her study correlated

strongly with age, weight and body surface area (BSA).

Among the derived parameters, Left ventricular mass

(LVM) was also highly correlated with these parameters

but Fractional Shortening (FS), Ejection Fraction (EF)

and LA/AO ratio showed a weak negative correlation.

The BSA correlated best with the echocardiographic

parameters studied, followed by height, weight and then

age. These findings were consistent with the observa-

with age, weight, height and body surface area

8

Oberhansli, Brandon, Lacourt et al studied a cohort of

2

1 normal babies sequentially by M-mode echocardi-

ography during the first year of life and correlated their

results with age, weight and length and found that the

end diastolic and end systolic diameters of the left ven-

tricle, the diameters of the left atrium (LA), the aorta

(

AO) as well as septal and left ventricular posterior wall

thickness (IVS and LVPW) followed a linear correlation

5

6

1

12

tions made by So₁l₃linger et al Henry et al and Roge,

ment (POD) and were matched for age and sex and so-

cioeconomic class.

Silverman, Hart.

There have also been reports o₅_-f₆_,₁c₄a_-₁r₅diac dimensions in

The Transthoracic echocardiographic studies were per-

formed on each using a Hewlett-Packard SONOS 500

machine and transducer with a frequency of 5MHz. A

baseline two-dimensional echocardiographic examina-

tion was carried out on each subject to ascertain normal

intracardiac segmental anatomy. The M-mode recording

was derived with the simultaneous recording of a 2D –

mapping to ensure precision in location and direction of

the M-mode cursor. Cardiac measurements were ob-

tained according to the recommendations of the commit-

tee on m-mode standa₆rdization of the America Society

sickle cell anaemia patients.

The findings are

similar to reports from normal patients. Balfour, Covitz,

5

Doris et al in a study comparing 124 patients with

HbSS referred to a tertiary centre in the USA with 78

healthy controls noted that the left ventricular and left

atrial dimensions in the patient group increased with

age. This study also found a linear relationship between

the cardiac dimensions namely the left atrial dimension,

left ventricular EDD, left ventricular wall dimension and

the left ventricular mass and the BSA.

1

of Echocardiography. These include aortic root dimen-

6

Similarly, Lester et al in 1990 evaluated 44 children

sion (AO), left atrial dimension (LA), left ventricular

end diastolic diameter [LVEDD], left ventricular end

systolic diameter [LVESD], interventricular septal thick-

ness [IVS], left ventricular posterior wall thickness

[LVPW] . The left ventricular ejection fraction (EF),

fractional shortening (FS) and left ventricular mass were

derived from the m-mode measurements. Haemoglobin

concentration was determined using oxy-haemoglobin

method.

with HbSS aged 2 months to 18 years using M-mode

echocardiography and reported the relationship between

cardiac dimensions, weight and BSA. The study found a

direct linear relationship between the left ventricular

EDD, left ventricular mass, aortic root dimensions, left

atrial dimensions and the weight of the patients. The

dimensions obtained in the study were also observed to

increase with age. Simila₁r₄ findings have also been docu-

mented by Covitz et al who found that the right and

left ventricular diastolic dimensions, left ventricular free

wall and septal thickness, left atrial dimension, and aor-

tic root dimensions were significantly and directly corre-

lated with BSA.

Height was measured to the nearest 0.5cm with the child

barefoot, standing erect with the heels together against

the wall, and looking st₁r₇aight ahead with the back

against the graduated wall . The height was read with a

wooden ruler resting on the scalp and agains₁t₇the wall.

Length was measured using an infantiometer. Subjects

were weighed standing barefoot, wearing only their un-

derware using a Seca® scale. Weight was read to the

nearest 0.1kg. The scale was calibrated with a standard

In all the above studies, a multivariate analysis using

these parameters were never carried out hence the need

for this study which is aimed at documenting the pa-

rameter which best correlate with echocardiographic

parameters.

th

weight after every 10 measu₁₇rement or whenever it was

moved from place to place. All the participants could

walk.

Subjects and methods

BSA was estimated using a body s₇urface area normo-

1

gram based on weight and height. Each subject and

This was a prospective, cross sectional and analytical

study carried out at the Lagos University Teaching Hos-

pital (LUTH), Idi-Araba, Nigeria as part of a large study

between May 2005 and October 2005. The subjects in-

cluded 60 paediatric patients attending the LUTH sickle

cell anaemia outpatient clinic and were consecutively

recruited. They had haemoglobin genotype ‘SS’ con-

firmed with haemoglobin electrophoresis (as docu-

mented in the file) and were aged 12 months to 15 years.

They were in steady state at the time of recruitment.

control also had venous blood sample taken for estima-

tion of haemoglobin level.

Data was analyzed using Microsoft Excel program sup-

plemented by Megastat statistical package. Mean, stan-

dard deviation and other parameters were generated as

necessary for continuous data. The subjects and controls

included were compared using student t-test for continu-

ous data, and chi-square test for discrete data. The Pear-

son correlation coefficient r was calculated to determine

the relationship between the cardiac measurements, age,

weight, height, body surface area, as well as haemoglo-

bin concentration. Univariate and multivariate correla-

tion analysis was used to study the relationships between

selected continuous sets of data. The coefficients of cor-

relation and associated p-values were derived. Statisti-

cal significance was set at p-value < 0.05.

Inclusion criteria for the controls included: Haemo-

globin AA, absence of congenital or acquired heart de-

fects, absence of respiratory or renal disease, and ab-

sence of protein energy malnutrition and haemoglobin

concentration of 10g/l or higher.

Patients with congenital or acquired heart disease, renal

disease and hypertension were excluded from the study.

Healthy controls were from the Community Health Out-

patient and Well baby clinics and healthy children at-

tending other clinics at the Paediatric outpatient depart-

Ethical clearance for the study was obtained from the

ethical committee of the Lagos University Teaching

Hospital and informed consent was sought from parent

or care givers of subjects and controls before enrolment

into the study.

5

7

Results

there was significant linear relationship between haemo-

globin level (Hb) and the following echocardiographic

parameters namely; AO, EDD, ESD, in the controls.

In addition, IVS, FS, EF, and LA/AO were inversely

related to haemoglobin level in the controls but the

strength of the relationship was not statistically signifi-

cant. Similarly a weak and non-significant inverse rela-

tionship was observed between most of the echocardio-

graphic measurements (LA, LA/AO, EDD, ESD,

LVPW, FS, EF, and LVM) and the haemoglobin level in

the subjects (r < - 0.1, p > 0.3).

A total of 120 children were recruited into the study. Of

this number, 60 were test subjects in the 12 months to 15

years age bracket (Mean = 95.41 + 49.06 months) con-

firmed to have genotype SS by haemoglobin electropho-

resis. They were all in steady state. The control group

consisted of 60 children within the same age bracket

(mean = 95.4 + 50.92 months) (t = 0.04, p = 0. 96) who

had haemoglobin genotype AA. The mean (SD) haemo-

globin level in the subjects was significantly lower than

in the controls (77.23 + 12.88g/L Vs 121 + 16.09g/L, p<

0

.001)

Multiple regression

Correlation of echocardiographic measurements with

weight, height, body surface area and age

The results show that height correlated significantly

with AO and ESD in subjects and controls. It also re-

mained significantly correlated to EDD and LVPW in

the subjects. While weight remained significantly corre-

lated to LA and LA/AO in the subjects and surprisingly

was significantly correlated to FS and EF in the controls.

However HB which did not achieve any significant cor-

relation with any of the echo parameters on univariate

analysis was found to be significantly correlated with

LVM. All the parameters (Height, Weight, BSA, HB)

except age were significantly correlated with LVM both

in univariate and multivariate analysis. Age that showed

significant correlation on univariate analysis to all the

direct echo parameters and LVM was only significantly

correlated with LVPW in subjects in the multiple

The Univariate correlation of echocardiographic meas-

urements with weight, height and body surface and age,

each shows that the echocardiographic measurements

namely the AO, LA, EDD, ESD, LVPW,IVS and the

LVM each increased progressively with body weight,

height, age and BSA in the study population (r > 0.3 and

p < 0.001). There was no significant association between

weight and FS, EF, and LA/AO in the study population

(

p > 0.1 ).

Haemoglobin concentration

In the correlation between the echocardiographic meas-

urements and haemoglobin concentration in the subjects,

regression models.(Table1 and 2).

Table 1: Multiple regression models with age, height, weight and haemoglobin level as independent variable and the

direct echo parameters as dependent variables in subjects.

2

Echo parameter

Age

HB

HT

WT

R

p

AO

t

p

0.12

0.91

0.07

0.94

2.40

0.02

-0.41

0.68

0.53

0.000000003

LA

p

t

0.41

-0.83

0.32

-1.01

0.33

0.98

0.02

3.28

1.33

0.58

0.0000000132

EDD

p

t

0.53

-0.64

0.32

-1.00

0.02

2.34

0.19

0.61 0.00000000118

ESD

t

p

0.84

0.40

-0.24

0.82

2.47

0.49

-0.04

0.02

0.40

0.97

0.0000151

LVPW t-2.01

p

0.20

0.04

2.34

0.84

0.20

0.02

0.003

0.63

IVS

t

p

-1.00

0.32

0.06

0.95

0.69

0.49

1.61

0.11

0.14

0.01

5

8

Table 2: Multiple regression models with age, height,

weight and haemoglobin level as independent variable

and the derived echo parameters as dependent variables

in subjects.

anthropometric indices i.e. body weight, height and BSA

in both subjects and contr₅o_,₆ls. This finding is similar to

those of earlier studies

.The functional parameters

(FS and EF) were independent of the anthropometric

indices in both groups. This implies that irrespective of

body dimensions, the percentage of blood ejected by the

ventricles and the degree of the myocardial contractility

with each heart best remains constant in order to main-

tain the normal body physiology.

Echo

parameter Age

2

Hb

HT WT

R

p

LA/AO t -0.67 -0.81 -0.80 2.25 0.03 0.24

P 0.51 0.42 0.43 0.03

Similarly, most of the echocardiographic parameters

showed a direct linear relationship with the age in both

groups. The correlation between age and echocardio-

FS

EF

t

P 0.47 0.57 0.21 0.24

0.73 -0.57 -1.28 1.19 0.12 0.03

5

graphic findings was also noted by Balfour et al and

1

5

t

P 0.56 0.42 0.24 0.21

0.59 -0.81 -1.19 1.28 0.02 0.57

Cipolotti et al . This linear increase of the cardiac di-

mensions with increasing age is in keeping with the nor-

mal age-d₁e₁pendent effect of growth and development on

the heart . However, the functional parameters (FS and

EF) in this study did n₈ot show a significant r₁e₉lationship

BSA correlated significantly with LA, EDD and LVPW

in subjects. However Hb which did not achieve any sig-

nificant correlation with any of the echo parameters on

univariate analysis was found to be significantly corre-

lated with LVM. All the parameters (height, weight,

BSA, HB) except age were significantly correlated with

LVM both in univariate and multivariate analysis.

1

with age. Pombo et al and Covarrubias et al here also

demonstrated that these parameters remain constant irre-

spective of age. This also follows same explanation as

above.

(

table 3)

There has been no agreement on which indicator of body

size (height, weight and BSA) or age should be used to

relate to and correct echocardiographic measurements.

An observation of a linear relationship of echocardio-

graphic measurements with height, weight, body surface

area and age has been found in this study using univari-

ate analysis. With multivariate analysis this study found

that the height corrsubjects. However, studies in normal

children have found elates best with the dimensions fol-

lowed by BSA in that the BSA correlates best with the

dimensions and also shows a linear relationship with the

measurements compared to the other anthropometric

Table 3: Multiple regression models with age in

months, body surface area (BSA) and haemoglobin level

(

rameters as dependent variables in subjects.

HB) as independent variable and the direct echo pa-

Echo

parameter Age BSA Hb

R

p value

t

p

t

p

t

p

9

, 10,

indices and the age . This difference may be ex-

plained by the fact that a multivariate analysis was car-

ried out in this study which was not done in the earlier

studies.

AO

1.49 1.87 0.24 0.51

.14 0.07 0.81

-0.63 4.22 -0.81 0.56

.53 0.0001 0.42

0.64 3.32 -0.74 0.66

.52 0.002 0.46

0.75 1.76 0.00 0.36

.46 0.08 0.99

-0.89 2.27 0.43 0.15

.38 0.03 0.67

-0.50 -1.84 1.91 0.16

0.00001

0.0065

0

LA

0

EDD

ESD

LVPW

IVS

0

Conclusion

0

Direct echocardiographic parameters in sickle cell anae-

mia patients are significantly correlated with anthropom-

etric indices (Height, weight, BSA) and age. Our data

suggest best correlation with height in these subjects and

significant correlation of haemoglobin level with left

ventricular mass in subjects and controls.

0.0048

Discussion

This study aimed to correlate the echocardiographic

measurements with the weight, height, age, body surface

area and haemoglobin level of study subjects and to

compare this with those of healthy age, sex and socio-

economic class matched controls using univariate and

multivariate analysis to be able to document which of

the anthropometric parameters best correlate with echo-

cardiographic parameters.

Conflict of interest: None

Funding: None

Acknowledgement

The cooperation of the study population, their parent

and caregivers is gratefully acknowledged. The support

of Chief Nursing officer M A Bastos is highly

appreciated

In this study, most of the direct echocardiographic pa-

rameters and the LVM increased progressively with

5

9

References

1

.

Francis JRB, Johnson CS. Vascu-

lar occlusion in sickle diseases:

Current concepts and unanswered

questions. Blood 1991; 77:1405-

8. Oberhansli I, Brandon G, Lacourt

G, Friedli B. Growth patterns of

cardiac structures and changes in

systolic time intervals in the new-

born and infant. Acta Paediatr

1980; 69:239-47.

9. Sutton MS, Picard MH, Oldershaw

PJ Saaccheti R, Gibson DG. Effect

of age related changes in chamber

size, wall thickness and heart rate

on left ventricular function in nor-

mal children. Br Heart J 1982; 48:

342-51.

14. Covitz W, Espland M, Gallagher D,

Hellen brand W, Leff S, Talner N.

The heart in Sickle cell anaemia

(the cooperative study of sickle

cell disease). Chest 1995;108:

1214-19

15. Cipolotti R, Costa GB, Lima

WHC, Franco RP, Mello EV,

Dalfabbro AL, Gurgel RQ, Cuevas

LE. Echocardiographic characteris-

tics of patients with sickle cell

anaemia in. Sergipe, Brazil. J Trop

Paediatr 20001; 47: 73-6.

1

4.

2

3

.

Odia OJ. Electrocardiography

observation in sickle cell anaemia

patients. Trop Cardiol 1990; 16:

1

35-37.

Val-Mejas J Lee WK, Weisse

ABR, Ojan TJ. Left Ventricular

performance during and after

sickle cell crisis. Am J Med 1979;

9

7: 585-591.

10. Oladokun REM, Omokhodion SI.

Left ventricular dimensions and

functional parameters in apparently

normal Nigerian Children. Cardiol-

ogy Tropicale 2002;28:65-71.

11. Sollinger R, Eible F, Minhask.

Echocardiography in the normal

neonate. Circulation 1973; 47: 108

-18

12. Henry WL, Gardin JM, Wane JH.

Echocardiographic measurements

in normal subjects from infancy to

old age. Circulation 1980; 62:

1054-60.

16. American Society of Echocardi-

ography. Recommendations for

continuous quality improvement in

echocardiography. J Am Soc Echo

1995; 8 [5] part 2 supplement.

17. Emodi KJ, Kaine WN. Weights,

heights and Quetelet’s indices of

children with sickle cell anaemia.

Niger J Paed 1996;23: 37-41.

18. Pombo JF, Troy BL, Russell RO.

Left ventricular volumes and ejec-

tion fraction by echocardiography.

Circulation 1971; 43:480-90.

19. Covarrubias E.A, Sheikh M.U,

Solanki D, Morjaria M, Fox LM.

Left ventricular function in sickle

cell anaemia: A non-invasive

evaluation. South Med J 1980; 83:

342-344.

4

.

Ogunkunle OO, Jaiyesimi F. Car-

diovascular findings in children

with sickle cell disease.

Niger J Paed 1992; 19: 37-40

Balfour IC, Covitz W, Daris H,

Rao PS., Strong WB, Alpert BS.

Cardiac size and function in chil-

dren with sickle cell anaemia Am

Heart J 1984; 108:345-50.

Lester LA, Sodt PC, Hutcheon N,

Arcilla RA. Cardiac abnormalities

in children with sickle cell anae-

mia. Chest 1990; 98: 1169-1174.

Halon DA, Amitai N, Gotsman

MS, Lewis BS. Serial echocardi-

ography in the first three months

of life. J Cardiol 1979; 9 [5]: 393-

5

.

6

7

.

13. Roge CLL, Silverman NH, Hart

PA, Ray RM. Cardiac structure

growth pattern determined by echo-

cardiography. Circulation 1978;

57:285-90.

4

04.