Metabolic Bone Disease in Preterm infants: Relationship between radiologic grading in the wrist and serum biochemical markers

1. Welcome
Shubhra Prakash Paul
MD (Pediatrics) Part III
Bangladesh Institute of Child
Health

2. Article Title
Metabolic bone disease in preterm
infants: Relationship between
radiologic grading in the wrist and
serum biochemical markers

3. Source
Authors
S. K. You, J. E. Lee , S. M. Lee, H.-H. Cho
Journal
Diagnostic and Interventional
Imaging
2017; 98(11): 785-791
DOI
10.1016/j.diii.2017.06.008

4. Journal Cover
Page
1st page of Article

5. Introduction
* Rehman MU, Narchi H. Metabolic bone disease in the preterm infant: current state and future directions. World J Methodol 2015;5:115—
21.
* Harrison CM, Gibson AT. Osteopenia in preterm infants. Arch Dis Child Fetal Neonatal Ed 2013;98:F272—5.
# Backstrom MC, Kuusela AL, Maki R. Metabolic bone disease of prematurity. Ann Med 1996;28:275—82.
• Preterm and low birth weight (LBW)
infants are at an increased risk of
developing metabolic bone disease
(MBD).*
• The prevalence of MBD in preterm infants
in 1996#
–VLBW, < 1500 g was 20–30%
–ELBW, < 1000 g was 50–60%

6. Introduction (contd.)
a. Done SL. Fetal and neonatal bone health: update on bone growth and manifestations in health and disease. Pediatr Radiol
2012;42:S158—76.
b. Fewtrell MS, Cole TJ, Bishop NJ, Lucas A. Neonatal factors predicting childhood height in preterm infants: evidence for a
persisting effect of early metabolic bone disease? J Pediatr 2000;137:668—73
• The fetus absorbs calcium and phosphorus
(P) mainly during the third trimester of
pregnancy; therefore, preterm infants lose
the opportunity to store minerals. a
• Even after correction of radiologic signs of
MBD and biochemical markers, MBD can
still have a negative influence on the linear
growth of affected infants. b

7. Introduction (contd.)
* Rustico SE, Calabria AC, Garber SJ. Metabolic bone disease of prematurity. J Clin Transl Endocrinol 2014;1:85—91
# Aiken C, Sherwood R, Lenney W. Role of plasma phosphate measurements in detecting rickets of prematurity and in monitoring treatment. Ann Clin
Biochem 1993;30:469—75.
# Kovar I, Mayne P, Barltrop D. Plasma alkaline phosphatase activity: a screening test for rickets in preterm neonates. Lancet 1982;319:308-10.
# Kelly A, Kovatch KJ, Garber SJ. Metabolic bone disease screening practices among US neonatologists. Clin Pediatr 2014;53:1077—83
• The definition of MBD varies, as patients
can present with osteopenia, osteoporosis,
rickets, and fractures.*
• Commonly used as serum biochemical
markers for the detection of MBD are#
–High levels of alkaline phosphatase
(ALP)
–Low levels of Phosphate

8. Introduction (contd.)
* Abrams SA, Committee on N. Calcium and vitamin D requirements of enterally fed preterm infants. Pediatrics
2013;131:e1676—83.
• The American Academy of Pediatrics
recommends carrying out long-bone (wrist
or knee) radiography to confirm the
diagnosis of rickets in preterm infants and
to follow-up these individuals at 5—6
week intervals until it is resolved.*

9. Objectives
• To evaluate
–The relationship between wrist
radiographs and levels of ALP or
phosphate.
–The standard biochemical markers for
MBD screening.
–The optimal cut-off values of
biochemical markers to predict MBD.
• To assess the optimal timing of wrist
radiographs.

10. Methodology(contd.)
Type of study
Retrospective cohort study
Place of study
NICU of Kyungpook National University
Medical Center, 807 Hoguk-ro, Buk-gu,
Daegu 41404, Republic of Korea
Duration of study
From January 2014 to September 2016
Study population
159 neonates

11. Methodology (contd.)
Inclusion Criteria
Infants born at less than 37 weeks of
gestational age
Exclusion Criteria
Congenital skeletal anomalies

12. Study Procedure (contd.)
Biochemical analysis
All preterm infants were measured for
standard biochemistry parameters (serum
ALP and Phosphate) weekly or bi-weekly
until discharge.
• The first measurements after birth, levels
on the same day of the first wrist
radiography
(ALP-s, P-s; if these data were not
recorded on the same day, the levels
recorded closest to that day were used
instead)

13. Study Procedure (contd.)
Biochemical analysis
• The highest ALP level before the first
radiography (ALP-hb) and during the
serial follow-up (ALP-h).
• The lowest P level before the first wrist
radiography (P-lb) and during the serial
follow-up period (P-l).

14. Study Procedure (contd.)
Radiological analysis
• The first and serial bilateral wrist
radiography were retrospectively reviewed
independently by two board-certified
pediatric radiologists who were blinded to
the clinical findings.
• The most severe grade recording during
the serial follow-up period was
determined in consensus by two
radiologists for analysis

15. Study Procedure (contd.)
Radiological analysis
• In the first analysis, comparison of the
biochemistry values for each of the four
grades.
• For the second analysis, the patients were
divided into two groups according to the
presence or absence of radiographic
changes.
• The presence of MBD was graded on he
basis of the wrist radiography according to
previous studies.*
* Koo WW, Gupta JM, Nayanar VV, Wilkinson M, Posen S. Skeletal changes in preterm infants. Arch Dis Child 1982;57:447—52..

16. Study Procedure (contd.)
Radiological Grading of
Osteopenia*
0 Normal epiphysis of the
radius or ulna (fig. a)
1 loss of the dense white
line, increased sub-
metaphyseal lucency,
and/or thinning of the
cortex (fig. b)
2 Irregularity, fraying,
splaying, and/or cupping
of the metaphysis (fig. c)
3 Fractures with grade 2 (fig.
d)* Koo WW, Gupta JM, Nayanar VV, Wilkinson M, Posen S. Skeletal changes in preterm infants. Arch Dis Child 1982;57:447—52..

17. Statistical Analysis
Software Used IBM SPSS Statistics for Windows
(Version 21.0., IBM Corp., Armonk, NY,
USA).
Statistical Analysis
 Statistical significance was defined as P <
0.05
 The mean and standard deviation were
calculated for the clinical and demographic
data of each group.
 In the first analysis, the biochemistry results
were compared among the four
radiological grades using one-way analysis
of variance (ANOVA) with a Tukey multiple-

18. Statistical Analysis
Statistical Analysis
 In the second analysis, the results were compared
between two groups according to the presence
and absence of radiographic changes using the
Student t-test.
 A receiver operator characteristic (ROC) curve was
constructed to determine the optimal cut-off
values of the biochemical markers for the
detection of MBD.
 The inter-observer reliability was assessed using
the intra-class correlation coefficient (ICC) which
was 0.94 (95% confidence interval [CI] = 0.92—
0.96, P < 0.001).

19. Result
Grade
Grade 0 Grade 1 Grade 2 Grade 3 P-value
n = 94 n = 39 n = 19 n = 7 Grade 0
vs.
grade 1,
2, 3
Grade 0
vs. 1 vs. 2
vs. 3
GA
(weeks)
30.5 ±
2.1
28.6 ±
2.0
27.3 ±
1.9
26.6 ±
2.1 < 0.001
BW
(gram)
1340.9 ±
262.1
1116.7 ±
219.4
940.0 ±
318.2
971.4 ±
313.9 < 0.001
LS
(weeks)
7.6 ±
2.5
10.1 ±
3.4
14.3 ±
6.5
23.8 ±
8.6 < 0.001
GA : Gestational Age; BW: Birth Weight; LS: Length of stay.
Table 1. Patient demographics and clinical findings
according to the radiological grade of MBD.

20. Result
Grade
Grade 0 Grade 1 Grade 2 Grade 3 P-value
ALP
(U/L) n = 94 n = 39 n = 19 n = 7
Comparison between
Grades
Grade 0/1 vs. Grade 2/3
ALP-s 415.8 ±
146.3
484.6 ±
148.3
679.1 ±
261.6
799.0 ±
231.2
< 0.001
ALP-hb 495.9 ±
140.9
552.0 ±
157.7
809.9 ±
379.5
1021.0 ±
534.7
< 0.001
ALP-h 512.6 ±
168.4
609.3 ±
192.1
917.4 ±
423.0
1694.5 ±
627.7
< 0.001
ALP-s: First measurements after birth; ALP-hb: Highest ALP level
before the first radiography; ALP-h: Highest ALP level during the serial
follow-up
Table 2. Comparison of biochemical markers among the
four radiological grades (alkaline Phosphatase [ALP])

21. Result
Grade
Grade 0 Grade 1 Grade 2 Grade 3 P-value
n = 94 n = 39 n = 19 n = 7
Comparison between
Grades
0/1 vs.
2/3
0 vs. 2 0 vs.
2/3
P-lb 3.0 ± 1.2 2.1 ± 0.
7
1.9 ±
0.6
1.9 ±
0.7
0.001
P-l 2.9 ± 1.2 2.3 ± 0.
7
1.8 ±
0.6
1.3 ±
0.7
< 0.001
/ 0.001
P-lb: The lowest P level before the first wrist radiography; P-l : The
lowest P level during the serial follow-up period
Table 2 (contd.): Comparison of biochemical markers
among the four radiological grades (Phosphate level [P])

22. Result
Characteristics
Radiographic findings
Normal Abnormal P-value
n = 94 n = 65
Gestational age
(weeks)
30.5 ± 2.1 28.0 ± 2.1 < 0.001
Birth weight
(grams)
1340.9 ± 262.1 1049.4 ± 270.
7
< 0.001
Length of stay
(weeks)
7.6 ± 2.5 12.8 ± 6.6 < 0.001
Table 3. Patient demographics and clinical findings
between groups with normal and abnormal radiographic
findings.

23. Result
Characteristics
Radiographic findings
Normal Abnormal P-value
n = 94 n = 65
ALP-s (U/L) 415.8 ± 146.3 575.3 ± 225.8 < 0.001
ALP-hb (U/L) 415.8 ± 146.3 675.2 ± 330.3 < 0.001
ALP-h (U/L) 512.6 ± 168.4 816.2 ± 471.9 < 0.001
P-s (mg/dL) 6.4 ± 3.9 4.3 ± 1.3 < 0.001
P-lb (mg/dL) 3.0 ± 1.2 2.2 ± 0.7 < 0.001
P-l (mg/dL) 2.9 ± 1.2 2.1 ± 0.8 < 0.001
Table 4. Comparison of biochemical markers between
groups with normal and abnormal radiographic findings.

24. Result
In this study
• Predictor of radiological MBD in the wrist
of preterm infants is the highest ALP level
detected at 6.9 ± 5.3 weeks after birth.
• Abnormal radiologic changes (grade 1—
3) were found at the mean ALP-h level of
816.2 U/L.
• Cut-off value of 816.2 U/L for ALP-h, the
sensitivity and specificity were 35.4% and
94.7%, respectively.

25. Result
Figure 2. Box plots of (a)
ALP-s, (b) ALP-hb, and (c)
ALP-h for the four
radiographic grades. The
median values of the ALP-s,
ALP-hb, and ALP-h levels of
the grades 2/3 groups were
significantly larger than
those of the grades 0/1
groups (P < 0.001).

26. Result
Figure 3. Box plots of (a) P-lb and (b) P-l: a: the median
values of P-lb for the grade 2 group were significantly
lower than those of the grade 0 group (P = 0.001); b: the
median values of P-l of grades 2/3 were significantly lower
than those in grade 0 (P< 0.001 and P = 0.001,
respectively).

27. Result
Figure 4. Receiver operator characteristic (ROC) curve for
ALP-h to discriminate metabolic bone disease (MBD) in
preterm infants.

28. Discussion
• In this study, ALP-s, ALP-hb, and ALP-h
levels of infants with grades 2/3 were
significantly higher than those in infants of
grades 0/1. P-lb and P-l levels of infants
with grades 2/3 were also significantly
lower than those in infants with grade 0.

29. Discussion
• Rustico et al. (2015) showed that only the
serum ALP level was increased in a fracture
group compared to an osteopenia group,
and there were no significant differences
in serum P, calcium, or parathyroid
hormone (PTH) levels.
• ALP level was not significantly higher in
ELBW infants with radiologically confirmed
rickets compared to those without rickets
(Mitchell et al. 2009) and VLBW infants
between radiologically confirmed
rickets/osteopenia and normal groups
(Arani et al. 2015 )

30. Discussion
For detecting osteopenia in preterm infants
–ALP level of >500 U/L was a useful
indicator of MBD (Kelly et al. 2014)
–Increased serum ALP levels (>700 U/L)
at 3 weeks after birth showed 73%
sensitivity and 73% specificity (Hung et
al. 2011)
–increased serum ALP level (>495 U/L) at
5 weeks after birth had 81% sensitivity
and 87% specificity (Lee et al. 2011)

31. Discussion
– For radiological detection of osteopenia in
preterm infants Moreira et al. (2014)
suggested Knee radiography where as
Tokuriki et al. (2016) suggested humerus
cortical bone thickness .
– Further studies comparing radiological
findings of the wrist and knee or comparing
those of the wrist/knee to those of the
humerus could be helpful to better
diagnose MBD in preterm infants.

32. Limitation
• No clear inclusion and exclusion criteria
had been mentioned in the article.
• Its difficult to quantify bone mineral density
based on wrist radiograph, comparison
with more specific test like DEXA (Dual
Energy Xray Absoptiometry) might yield
better significance.

33. Limitation
Author declared limitations
• As it was a retrospective study, results
might have been influenced by the
heterogeneity of the study population,
which included both VLBW and ELBW
infants.
• Second, a small number of patients were
included.
• Other laboratory findings such as serum
calcium, PTH, and 25-hydroxyvitamin D
concentrations were not analyzed.

34. Conclusion
Taking the wrist radiography with
reference to an ALP level measured at
around 6.9 weeks after birth could be
helpful for screening of MBD in
preterm infants .