ORIGINAL ARTICLE

G. Perinetti Maturation of the middle phalanx

L. Perillo
L. Franchi of the third finger and cervical
R. Di Lenarda vertebrae: a comparative and
L. Contardo
diagnostic agreement study

Authors' affiliations: Perinetti G., Perillo L., Franchi L., Di Lenarda R., Contardo L. Maturation
G. Perinetti, R. Di Lenarda, L. Contardo, of the middle phalanx of the third finger and cervical vertebrae: a
Department of Medical, Surgical and
Health Sciences, School of Dentistry,
comparative and diagnostic agreement study
University of Trieste, Trieste, Italy Orthod Craniofac Res 2014; 17: 270–279. © 2014 John Wiley & Sons A/S.
L. Perillo, Department of Orthodontics, Published by John Wiley & Sons Ltd
School of Dentistry, Second University of
Naples, Naples, Italy
L. Franchi, Department of Orthodontics, Structured Abstract
School of Dentistry, University of Objective – Diagnostic agreement on individual basis between the third
Florence, Florence, Italy middle phalanx maturation (MPM) method and the cervical vertebral
L. Franchi, Department of Orthodontics
maturation (CVM) method has conjecturally been based mainly on overall
and Pediatric Dentistry, School of
Dentistry, The University of Michigan, correlation analyses. Herein, the true agreement between methods
Ann Arbor, MI, USA according to stage and sex has been evaluated through a comprehen-
sive diagnostic performance analysis.
Correspondence to:
G. Perinetti
Subjects and methods – Four hundred and fifty-one Caucasian subjects
Struttura Complessa di Clinica Odontoia- were included in the study, 231 females and 220 males (mean age,
trica e Stomatologica 12.2  2.5 years; range, 7.0–17.9 years). The X-rays of the middle
Ospedale Maggiore
phalanx of the third finger and the lateral cephalograms were examined
Piazza Ospitale 1
34129 Trieste for staging by blinded operators, blinded for MPM stages and subjects’
Italy age. The MPM and CVM methods based on six stages, two pre-pubertal
E-mail: [email protected] (1 and 2), two pubertal (3 and 4), and two post-pubertal (5 and 6), were
considered. Specifically, for each MPM stage, the diagnostic
performance in the identification of the corresponding CVM stage was
described by Bayesian statistics.
Results – For both sexes, overall agreement was 77.6%. Most of the
disagreement was due to 1 stage apart. Slight disagreement was seen for
the stages 5 and 6, where the third middle phalanx shows an earlier
maturation.
Conclusions – The two maturational methods show an overall satisfacto-
rily diagnostic agreement. However, at post-pubertal stages, the middle
phalanx of the third finger appears to mature earlier than the cervical
vertebrae. Post-pubertal growth phase should thus be based on the
Date:
Accepted 14 August 2014
presence of stage 6 in MPM.

DOI: 10.1111/ocr.12052
Key words: cervical vertebrae; diagnosis; finger phalanges; growth
© 2014 John Wiley & Sons A/S.
Published by John Wiley & Sons Ltd
 Perinetti et al. MPM and CVM agreement

Introduction Materials and methods

Study population and design
When dealing with skeletal disharmonies for
treatment efficiency reasons, the precise identifi- The databases between January 2008 and August
cation of skeletal maturity, that is, the growth 2013 of the Sections of Stomatology of the
phase, with particular regard to the onset of the Department of Medical, Surgical and Health Sci-
pubertal growth spurt, is required (1, 2). Several ences, University of Trieste, and of the Depart-
indices have been proposed to identify the skele- ment of Oral Sciences, Second University of
tal maturation phases (1–6). The most com- Naples, were screened. This study included sub-
monly used are the radiography-based, hand- jects who were seeking orthodontic treatment.
wrist maturation [for review, see (7)] and cervical Signed informed consent was obtained from the
vertebral maturation (CVM) [for review, see (1)] parents of the subjects prior to study entrance,
methods. and the protocol was reviewed and approved by
Several studies have previously evaluated the the local ethical committee. In particular, an
relationship between these two methods [for X-ray of the middle phalanx of the middle finger
review, see (8)] generally reporting a high and a lateral cephalogram is taken as part of the
degree of correlation. However, all of these pre- routine clinical recording. The following inclu-
vious studies were hampered by the use of a sion criteria were applied: 1) age between 7 and
specific CVM recording (9) not consistent with 18 years; 2) absence of anomalies of either the
the described methods (1, 10) validated in clini- fingers or the vertebrae; 3) good general health
cal trials (11) or by lack of an analysis of diag- with the absence of any nutritional problems; 4)
nostic agreement in individual subjects (8, 12). no history of trauma at the cervical region or
Indeed, a high correlation coefficient does not right hand; and 5) Caucasian ethnicity. A total
necessarily prove a diagnostic agreement in of 451 subjects (231 females and 220 males)
individual subjects, as was recently showed for were included in the study (mean age,
dental maturation (13). This issue may be 12.2  2.5 years; range, 7.0–17.9 years). In a pos-
addressed by a dedicated diagnostic perfor- teriori power analysis, sample size of 278 sub-
mance analysis that is, however, still missing. jects is enough to detect an agreement between
Moreover, very few studies (14–16) have specifi- the maturational methods as low as 30% consid-
cally been focused on the correlations between ering a relative error (the difference between the
the middle phalanx maturation (MPM) of the estimated and true reliability), as low as 20%
third finger and the CVM method. The results with a power of 0.8, and an alpha set at 0.05
of these investigations were further limited by (17).
the lack of an accurate recording of the repeat-
ability of the measurements of both matura- Radiographic recordings
tional methods (14), or because only male
subjects were included (15). The radiographic recording of the middle
This study was designed to address the fol- phalanx of the middle finger was performed as
lowing issues: 1) Does the middle phalanx of previously reported (18). Briefly, the patients
the third finger and cervical vertebral matura- were instructed to place their right hand with
tions have satisfactorily diagnostic agreement? the palm downward on a flat table and with the
and 2) If disagreement is seen, how is this third finger straight and centered on a standard
structured among the different stages or sexes? 3 9 4 cm periapical sensor (Du € rr Dental, Bietig-
This study ultimately verified whether the MPM heim-Bissingen, Germany). The cone of the den-
method, as proposed herein, may be proposed tal X-ray machine (Kodak 2200 intraoral x-ray
as a valid indicator of growth phase in individ- system; Eastman Kodak Company, Rochester,
ual subjects. NY, USA) was positioned in light contact with

Orthod Craniofac Res 2014;17:270–279 | 271

Perinetti et al. MPM and CVM agreement

the middle phalanx and perpendicular to the MPS2

dental X-ray sensor. Settings were of 70 kV and When the epiphysis is at least as wide as the
7 mA with an exposure time of 0.097 s. An auto- metaphysis (5) with sides increasing thickness
matic developer (VistaScan PERIO; Du € rr Dental) and showing a clear line of demarcation at right
was used for film processing. A dedicated X-ray angle (5). In case of asymmetry between the two
machine (KODAK 8000C; Eastman Kodak Com- sides, that is, one typical of MPS2 and the other
pany) was employed for the recording of lateral less mature, the former is used to assign the
head cephalogram. Settings were of 73–77 kV, stage. This stage was earlier reported as SMI2 (3)
12 mA, with an exposure time of 0.80 s. Radio- or as MP3-FG described to be attained 1 year
graphs of low quality were excluded. before the onset of the pubertal growth spurt
(5).
Middle phalanx maturation (MPM) method
MPS3
The MPM method as proposed herein comprises When the epiphysis is either as wide as or wider
6 stages [middle phalanx stages, (MPS)], as than the metaphysis (5) with lateral sides show-
shown in Fig. 1. Definitions of the stages were ing an initial capping toward the metaphysis
based on previous descriptions by Fishman (3), (5). In case of asymmetry between the two
Ha€gg and Taranger (5), and Rajagopal and sides, for example, one typical of MPS3 and the
Kansal (14), with modifications: other less mature, the former is used to assign
the stage. Epiphysis and metaphysis are not
MPS1 fused. This stage was earlier reported as
When the epiphysis is narrower than the me- SMI6 (3) or as MP3-G, both described to be
taphysis, or when the epiphysis is as wide as attained at coincidence of the pubertal growth
metaphysis (5), but with both tapered and spurt (5).
rounded lateral borders (14) (Fig. 1, MPS1b).
Epiphysis and metaphysis are not fused. MPS4
This stage was earlier reported as MP3-F and When the epiphysis begins to fuse with the me-
described to be attained more than 1 year taphysis (5) although contour of the former is
before the onset of the pubertal growth spurt still clearly recognizable. Both sides of the
(5). epiphysis form obtuse angle to distal border,

Fig. 1. The third middle phalanx

(upper) and cervical vertebral
(lower) maturational stages. MPS,
third middle phalanx matura-
tional stage; CS, cervical vertebral
maturational stage.

272 | Orthod Craniofac Res 2014;17:270–279

 Perinetti et al. MPM and CVM agreement

and the capping is still clearly detectable. This CS3

stage was earlier reported as MP3-H and When the lower borders of both C2 and C3 have
described to be attained after the pubertal concavities, and the bodies of C3 and C4 are
growth spurt, that is, during the deceleration of either trapezoid or rectangular horizontal in
the curve of growth (5). shape. This stage has been reported to occur in
coincidence with the onset of the pubertal growth
MPS5 spurt, that is, acceleration curve of growth.
When the epiphysis is mostly, but not com-
pletely fused with the metaphysis (5), and the CS4
distal contour of the former begins to be less When the lower borders of C2–C4 have concavi-
clearly recognizable. This specific stage was ini- ties, and the bodies of both C3 and C4 are rectan-
tially proposed by Rajagopal and Kansal (14) and gular horizontal. This stage has been described to
reported as MP3-HI and was reported to be be attained at coincidence of the pubertal growth
attained toward the end of the pubertal growth spurt, but after the peak height velocity, that is,
spurt (14). during the deceleration curve of growth.

MPS6 CS5
When the epiphysis totally fused with the me- When the lower borders of C2–C4 have concavi-
taphysis (5), and the distal contour of the former ties, and at least one or both of the bodies of C3
is not recognizable. This stage was earlier and C4 is square. This stage has been reported
reported as SMI10 (3) or as MP3-I, both to occur 1 year after the growth spurt.
described to be attained at the end of the puber-
tal growth spurt (5). CS6
An experienced orthodontist (GP), who was When the lower borders of C2–C4 have concavi-
blinded to the CVM stages, assessed the MPM ties, and at least one or both of C3 and C4 are
stages. rectangular vertical. This stage has been reported
to occur at least 2 years after the growth spurt.
Cervical vertebral maturation method The lateral cephalograms were cropped to
include C2–C4 and to eliminate any additional
The CVM method as initially proposed by Hassel information, such as stage of dentition that
and Farman (10), and subsequently modified might have biased the staging. An experienced
according to Baccetti et al. (1), comprises 6 orthodontist (LC) with 5-year experience in
stages (CS) as shown in Fig. 1 and as briefly the CVM method including training with the
defined as follows: developers of this staging, blinded to the MPM
stages and subjects’ age, assessed the CVM
CS1 stages.
When the lower borders of the second, third,
and fourth vertebrae (C2, C3, and C4) are flat, Statistical analysis
and the bodies of C3 and C4 are trapezoid in
shape. This stage has been reported to be All these analyses were performed for each sex
attained at least 2 years before the pubertal separately as well as for the whole sample. Mean
growth spurt. ages of the subjects, clustered according to each
stage of either maturational methods, have been
CS2 plotted, and within each MPM stage, the preva-
When only the lower border of C2 is concave, lence of the CVM stages was calculated. To deter-
and the bodies of C3 and C4 are trapezoid. This mine the degree of correlation between the two
stage has been reported to be attained 1 year maturational indices, the Spearman rank correla-
before the growth spurt. tion coefficient was used. The diagnostic agree-

Orthod Craniofac Res 2014;17:270–279 | 273

Perinetti et al. MPM and CVM agreement

ment between the MPM and CVM methods was

assessed by a linearly weighted j coefficient (19).
Moreover, a dedicated diagnostic performance
analysis, that is, Bayesian statistics, was also per-
formed to establish the diagnostic performance
of each MPM stage for the identification of each
corresponding CVM stage. This analysis included
sensitivity, specificity, positive and negative pre-
dictive values, accuracy, and positive likelihood
ratio (LHR) (20). A threshold of a positive LHR
of ≥10 (21) was considered for assessment of sat-
isfactory reliability of any MPM stage for the
identification of any of the CVM stages, that is,
satisfactory diagnostic agreement.
The percentage agreement and weighted j sta-
tistics were calculated for evaluation of the
intra-examiner agreement. For appraisal of the
stages of MPM and CVM, the intrarater weighted
j coefficients calculated on 30 pairs of record-
ings randomly selected were >0.92.
SPSS software 13.0 (SPSSâ Inc., Chicago, IL,
USA), MedCalcâ software 12.3.3.0 (MedCalc Soft-
ware, Mariakerke, Belgium) and the interactive
Stats Calculator (http://ktclearinghouse.ca/cebm/ Fig. 2. Chronological ages among the different third finger
toolbox/statscalc) were used to perform the sta- middle phalanx and cervical vertebral maturational stages for
females (upper) and males (lower). Data are presented as
tistical analyses. A p value < 0.05 was considered mean  standard error of the mean. MPS, third middle pha-
as significant. lanx maturational stage; CS, cervical vertebral maturational
stage. Females, n = 231; males, n = 220.

Results was seen. The total agreements were 77.9 and

77.3% for females and males, respectively. The
The comparative mean ages of the subjects for correlation coefficient between the two matura-
each MPM or CVM stage according to the sexes tional methods was 0.953 (p < 0.001) for the
are shown in Fig. 2. Mean ages were very similar whole sample, and of 0.953 (p < 0.001) and 0.952
with few exceptions for males at stage 5, in which (p < 0.001) for females and males, respectively.
the differences were about 0.6 years. For both the The weighted j coefficient (95% CI) for the diag-
maturational methods, the differences in chrono- nostic agreement among the different MPM and
logical ages between two consecutive stages from CVM stages was 0.88 (0.85–0.90) for the whole
2 to 5 ranged from about 0.6 to 1.5 years for both sample, and of 0.88 (0.84–0.91) and 0.87 (0.84–
sexes. Irrespective of the maturational method, 0.91) for females and males, respectively.
females attained stages 2–6 generally 1 year ear- Detailed relative distributions of the different
lier than males. Clinical examples for each MPM MPM stages according to CVM stages for
and CVM stages are shown in Fig. 3. females and males are summarized in Table 1.
Of the whole sample, 350 subjects (77.6%) The percentage of exact agreement of the MPM
showed a full agreement between the two matu- stages with the corresponding CVM stages
rational stages, 89 subjects (19.7%) showed a ranged between 68.3% (MPS2/CS2) and 97.6%
one-stage-apart disagreement, while in only 12 (MPS1/CS1) for females, and between 57.7%
cases (2.7%), a two-stage-apart disagreement (MPS5/CS5) and 90.6% (MPS1/CS1) for males.

274 | Orthod Craniofac Res 2014;17:270–279

 Perinetti et al. MPM and CVM agreement

Fig. 3. Clinical examples from six

subjects of this study for the third
middle phalanx and cervical
vertebral maturational stages.
MPS, third middle phalanx matu-
rational stage; CS, cervical verteb-
ral maturational stage. Note that
pubertal middle phalanx matura-
tion stages 3 and 4 may or may
not show undulation of the bor-
der of the metaphysis.

Table 1. Relative distributions of the different third middle phalanx maturational stages according to cervical vertebral
maturational stages for females and males

Cervical vertebral maturational stage

Third finger middle
Sex phalanx maturational stage CS1 CS2 CS3 CS4 CS5 CS6 Total

Females MPS1 97.6% (41) 2.4% (1) – – – – 42

MPS2 24.4% (10) 68.3% (28) 4.9% (2) 2.4% (1) – – 41
MPS3 3.1% (1) 15.6% (5) 75.0% (24) 6.3% (2) – – 32
MPS4 – – 13.5% (5) 78.4% (29) 8.1% (3) – 37
MPS5 – – 5.1% (2) 15.4% (6) 76.9% (30) 2.6% (1) 39
MPS6 – – – 5.0% (2) 25.0% (10) 70.0% (28) 40

Males MPS1 90.6% (58) 21.1% (6) – – – – 64

MPS2 21.1% (8) 73.7% (28) 5.3% (2) – – – 38
MPS3 – 6.5% (2) 77.4% (24) 16.1% (5) – – 31
MPS4 – – 7.1% (2) 75.0% (21) 10.7% (3) 7.1% (2) 28
MPS5 – – 7.7% (2) 26.9% (7) 57.7% (15) 7.7% (2) 26
MPS6 – – – 6.1% (2) 21.2% (7) 72.7% (24) 33

MPS, third middle phalanx maturational stage; CS, cervical vertebral maturational stage.
Data are presented as percentage (n) cases of each MPS within each CS. –, no cases. Females, n = 231; males, n = 220.

The other diagnostic performance parameters 90% for both the sexes. Finally, positive LHRs
of different MPM stages and the corresponding ranged between 12.5 (MPS2/CS2) and 131.3
CVM stages are summarized in Table 2. Sensitiv- (MPS1/CS1) for females, and between 10.7
ity ranged between 72.5% (MPS4/CS4) and (MPS5/CS5) and 22.5% (MPS1/CS1) for males.
96.6% (MPS6/CS6) for females, and between
60.0% (MPS4/CS4 and MPS5/CS5) and 87.9%
(MPS1/CS1) for males. Specificity values were all Discussion
above 90% for females and males.
Positive predictive values ranged between The present study reported on the diagnostic
68.3% (MPS2/CS2) and 97.6% (MPS1/CS1) for agreement between the different stages of matu-
females, and between 57.7% (MPS5/CS5) and ration of the middle phalanx of the third finger
90.6% (MPS1/CS1) for males. Negative predictive and the cervical vertebral on a population of
values and the accuracy values were all above Caucasic growing subjects.

Orthod Craniofac Res 2014;17:270–279 | 275

276 |
Table 2. Overall diagnostic performance of each third finger middle phalanx maturation stages for identification of corresponding the cervical vertebral maturational stages
for females and males
Perinetti et al. MPM and CVM agreement

Diagnostic parameter

Orthod Craniofac Res 2014;17:270–279

Third finger middle phalanx Positive predictive Negative predictive
Sex maturational stage Sensitivity Specificity value value Accuracy Positive LHR

Females MPS1 78.8 (64.3–88.5) 99.4 (96.8–99.9) 97.6 (87.6–99.6) 94.2 (89.9–96.7) 94.8 (91–97) 131.3 (19.8–869.9)
MPS2 82.4 (66.5–91.7) 93.4 (89–96.1) 68.3 (53–80.4) 96.8 (93.2–98.5) 91.8 (87.3–94.8) 12.5 (7.2–21.6)
MPS3 72.7 (55.8–84.9) 96 (92.3–98) 75 (57.9–86.7) 95.5 (91.7–97.6) 92.6 (88.3–95.4) 18.2 (8.9–37.2)
MPS4 72.5 (57.2–83.9) 95.8 (91.9–97.9) 78.4 (62.8–88.6) 94.3 (90.1–96.8) 91.8 (87.3–94.8) 17.3 (8.6–35)
MPS5 69.8 (54.9–81.4) 95.2 (91.1–97.5) 76.9 (61.6–87.3) 93.2 (88.7–96) 90.5 (85.8–93.8) 14.5 (7.4– 28.2)
MPS6 96.6 (82.9–99.4) 94.1 (90–96.6) 70.0 (54.6–81.9) 99.5 (97.1–99.9) 94.4 (90.5–96.8) 16.4 (9.4–28.6)

Males MPS1 87.9 (77.7–93.8) 96.1 (91.8–98.2) 90.6 (81–95.6) 94.9 (90.2–97.4) 93.6 (89.4–96.2) 22.5 (10.2–49.5)
MPS2 77.8 (61.9–88.3) 94.6 (90.3–97.0) 73.7 (58.0–85.0) 95.6 (91.6–97.8) 91.8 (87.2–94.8) 14.4 (7.7–27)
MPS3 80.0 (62.7–90.5) 96.3 (92.6–98.2) 77.4 (60.2–88.6) 96.8 (93.2–98.5) 94.1 (90.0–96.6) 21.6 (10.2–45.6)
MPS4 60.0 (43.6–74.4) 96.2 (92.4–98.1) 75.0 (56.6–87.3) 92.7 (88.1–95.6) 90.5 (85.6–93.8) 15.8 (7.3–34.3)
MPS5 60.0 (40.7–76.6) 94.4 (90.2–96.9) 57.7 (39.0–74.5) 94.8 (90.7–97.1) 90.5 (85.6–93.8) 10.7 (5.5– 20.7)
MPS6 85.7 (68.5–94.3) 95.3 (91.3–97.5) 72.7 (55.8–84.9) 97.9 (94.7–99.2) 94.1 (90.0–96.6) 18.2 (9.5–35)

MPS, third middle phalanx maturational stage; LHR, likelihood ratio.

Data are presented as mean (95% confidence interval). Females, n = 231; males, n = 220.
 Perinetti et al. MPM and CVM agreement

The CVM method has been correlated with both Moreover, the weighted j coefficients retrieved
the statural and the mandibular growth spurt (22, herein were very high up to 0.88 denoting a very
23), and even with levels of biomarkers of growth good overall agreement between the two matu-
(24, 25). A randomized clinical trial on functional rational methods. Again, this analysis was
treatments has proved the validity of this method missed in the previous studies (14–16).
in terms of skeletal outcome (11). Of note, previ- In the present study, about 78% of agreement
ous studies reported a low to good reproducibility was found, and disagreements showed mostly a
of the CVM method with weighted j values rang- single stage apart. The disagreement seen herein
ing from 0.36 to 0.79 according to the different may be explained by the fine transitional mor-
raters (26). A later investigation, using the same phological changes in either third middle pha-
sample and raters, reported that the assignment lanx or cervical vertebrae. However, good
of the shape of the bodies of C3 and C4 is the least agreements between the maturational methods
reproducible part of the CVM staging (27). In spite were those for stages 3, in which all the diagnos-
of this evidence, the intrarater reproducibility tic parameters showed a high performance of
obtained in the present study was satisfactory the MPM method in the identification of the
with weighted j of 0.92. The high reproducibility corresponding CVM staging (Table 2). Therefore,
seen herein was likely due to the extensive train- events responsible for the onset of the pubertal
ing of the rater. growth spurt, that is, hormonal changes (25),
In addition to specific training, the CVM would induce concomitant morphological
method also requires a lateral head film, which changes in both the third middle phalanx and
is available as a pre-treatment record. However, cervical vertebrae.
in several instances, optimal treatment timing is Satisfactory diagnostic agreement with accu-
to be delayed until after the diagnosis, making racy and positive LHRs values above 90% and
necessary a later re-evaluation of the growth 10%, respectively, for each MPM stage in the
phase. Moreover, the cervical vertebrae might be identification of the corresponding CVM stage
partially covered by the protection collar, which (Table 2). However, the calculation of the accu-
would be necessary to reduce radiation exposure racy and the positive LHRs takes into account
(28). Even though the radiographical recording both the identification of true positive and true
of the hand and wrist has been shown to be negative cases. Therefore, when dealing with
safer in terms of radiation exposure (28), this several possible clustering, an important diag-
method requires anyway additional X-ray expo- nostic parameter is the positive predictive value
sure of a hand and wrist as a whole, other than that gives an indication of the capability of a
a dedicated X-ray machine. Besides, re-execution given MPM stage in the identification of the cor-
of either a lateral head cephalogram or a hand- responding CVM stage, irrespective of the num-
and-wrist film for a re-evaluation of growth ber of true negative cases belonging to the other
phases is not indicated according to the most stages. By analyzing the positive predictive val-
recent guidelines (29). ues, in combination with the frequency distribu-
In the present study, the mean chronological tions of the maturational staging, a general
ages at which both females and males reached tendency for the MPM to reach the stage 6
the pubertal growth spurt, as recorded by the slightly earlier than the CVM is evident
MPS3 or CS3 (Fig. 2), are comparable with previ- (Table 2). The concept that small morphological
ously reported data (3, 5, 30). changes at the third middle phalanx may be bet-
The correlation coefficient seen in the present ter detected than those at the cervical vertebrae
study between the MPM and CVM methods is may also be responsible for this evidence. There-
very similar to that above 0.94 obtained in a pre- fore, from a clinical perspective, a safe diagnosis
vious investigation (16) that was based on the of the attainment of the post-pubertal growth
maturation staging of the middle phalanx of the phase especially in males should rely on the
third finger proposed by Ha €gg and Taranger (5). attainment of MPS6, rather than MPS5.

Orthod Craniofac Res 2014;17:270–279 | 277

Perinetti et al. MPM and CVM agreement

Further studies on the correlation/diagnostic middle phalanx appears to mature earlier than
performance of the present MPM method with the cervical vertebrae.
statural or mandibular growth, or even to other
hand-wrist maturation methods, are warranted.
Clinical relevance
Clinical implications
Individual monitoring of the growth phase, with
The availability of a radiographical method particular regard to the onset of the pubertal
based on a very minimal radiation exposure growth spurt, has been advocated to obtain pre-
appears to be a valuable tool in clinical practice. dictable treatment effects when dealing with
The MPM method appears to be a valid indica- skeletal malocclusions. Although slight differ-
tor of the onset of the pubertal growth spurt in ences exist, when compared with the cervical
individual subjects and may therefore find wide vertebral method, the maturational staging of
applications for planning treatment timing for the middle phalanx of the third finger appears to
functional treatments for skeletal class II or III be a valid indicator of the onset and of the end
(1) and constricted maxilla (31). Finally, the of the pubertal growth spurt in individual sub-
MPM method is of easy execution and interpre- jects. When a lateral head film is not available or
tation and may be performed in any clinical set- not clear in the cervical area, the middle phalanx
ting with minimal instrumentation. This method maturation method may be used as a valid alter-
may also be complementary when the CVM native method.
staging would be uncertain or not derivable form
a lateral cephalogram.
Competing interest

Conclusions The authors declare that they have no compet-

ing financial interests.
1) The MPM and CVM methods show an overall
Acknowledgements: The authors are deeply grateful
satisfactorily diagnostic agreement; 2) good  Favaretto (University of Trieste) and Dr.
to Dr. Nicolo
agreement for stages 3 that corresponds to the Serena Cavuoti (Second University of Naples) for their
onset of the pubertal growth spurt; and 3) a support in screening the radiographical records.
slight disagreement at stage 5, in which the third

References 3. Fishman LS. Radiographic evaluation 7. Flores-Mir C, Nebbe B, Major PW.

1. Baccetti T, Franchi L, McNamara of skeletal maturation. A clinically Use of skeletal maturation based on
JAJ. The cervical vertebral matura- oriented method based on hand- hand-wrist radiographic analysis as
tion (CVM) method for the assess- wrist films. Angle Orthod 1982;52:88– a predictor of facial growth: a sys-
ment of optimal treatment timing in 112. tematic review. Angle Orthod
dentofacial orthopedics. Semin Or- 4. Greulich WW, Pyle SI. Radiographic 2004;74:118–24.
thod 2005;11:119–29. Atlas of Skeletal Development of the 8. Santiago RC, de Miranda Costa LF,
2. Petrovic A, Stutzmann J, Lavergne J. Hand and Wrist, 2nd edn. Stanford, Vitral RW, Fraga MR, Bolognese AM,
Mechanism of craniofacial growth CA: Stanford University Press; 1959. Maia LC. Cervical vertebral matura-
and modus operandi of functional 5. Ha€gg U, Taranger J. Maturation tion as a biologic indicator of skele-
appliances: a cell-level and cyber- indicators and the pubertal growth tal maturity. Angle Orthod
netic approach to orthodontic deci- spurt. Am J Orthod 1982;82: 2012;82:1123–31.
sion making. In: Carlson DS, editor. 299–309. 9. Beit P, Peltomaki T, Schatzle M,
Craniofacial Growth Theory and 6. Perinetti G, Baccetti T, Contardo L, Signorelli L, Patcas R. Evaluating the
Orthodontic Treatment Monograph Di Lenarda R. Gingival crevicular agreement of skeletal age assess-
23 Craniofacial Growth Series. Ann fluid alkaline phosphatase activity as ment based on hand-wrist and cervi-
Arbor, MI: Center for Human a non-invasive biomarker of skeletal cal vertebrae radiography. Am J
Growth and Development, Univer- maturation. Orthod Craniofac Res Orthod Dentofacial Orthop
sity of Michigan; 1990. pp. 13–74. 2011;14:44–50. 2013;144:838–47.

278 | Orthod Craniofac Res 2014;17:270–279

 Perinetti et al. MPM and CVM agreement

10. Hassel B, Farman AG. Skeletal matu- 18. Abdel-Kader HM. The reliability of (IGF-I) testing. Am J Orthod Dento-
ration evaluation using cervical ver- dental x-ray film in assessment of facial Orthop 2008;134:209–16.
tebrae. Am J Orthod Dentofacial MP3 stages of the pubertal growth 26. Gabriel DB, Southard KA, Qian F,
Orthop 1995;107:58–66. spurt. Am J Orthod Dentofacial Ort- Marshall SD, Franciscus RG, Sout-
11. Martina R, Cioffi I, Galeotti A, Ta- hop 1998;114:427–9. hard TE. Cervical vertebrae matura-
gliaferri R, Cimino R, Michelotti A 19. Brenner H, Kliebsch U. Dependence tion method: poor reproducibility.
et al. Efficacy of the Sander bite- of weighted kappa coefficients on Am J Orthod Dentofacial Orthop
jumping appliance in growing the number of categories. Epidemi- 2009;136:478 e471–477; discussion
patients with mandibular retrusion: ology 1996;7:199–202. 478–480.
a randomized controlled trial. Or- 20. Greenhalgh T. Advertisements for do- 27. Nestman TS, Marshall SD, Qian F,
thod Craniofac Res 2013;16:116–26. nepezil. More convincing evidence of Holton N, Franciscus RG, Southard
12. Uysal T, Ramoglu SI, Basciftci FA, efficacy needs to be cited. BMJ TE. Cervical vertebrae maturation
Sari Z. Chronologic age and skeletal 1997;315:1623; author reply 1624. method morphologic criteria: poor
maturation of the cervical vertebrae 21. Deeks JJ, Altman DG. Diagnostic reproducibility. Am J Orthod Dento-
and hand-wrist: is there a relation- tests 4: likelihood ratios. BMJ facial Orthop 2011;140:182–8.
ship? Am J Orthod Dentofacial Ort- 2004;329:168–9. 28. Patcas R, Signorelli L, Peltomaki T,
hop 2006;130:622–8. 22. Franchi L, Baccetti T, McNamara JA Schatzle M. Is the use of the cervical
13. Perinetti G, Contardo L, Gabrieli P, Jr. Mandibular growth as related to vertebrae maturation method justi-
Baccetti T, Di Lenarda R. Diagnostic cervical vertebral maturation and fied to determine skeletal age? A
performance of dental maturity for body height. Am J Orthod Dentofa- comparison of radiation dose of two
identification of skeletal maturation cial Orthop 2000;118:335–40. strategies for skeletal age estimation.
phase. Eur J Orthod 2012;34:487–92. 23. Soegiharto BM, Moles DR, Cunning- Eur J Orthod 2012;35:604–9.
14. Rajagopal R, Kansal S. A comparison ham SJ. Discriminatory ability of the 29. Isaacson KG, Thom AR, Horner K,
of modified MP3 stages and the cer- skeletal maturation index and the Whaites E. Orthodontic Radiographs
vical vertebrae as growth indicators. cervical vertebrae maturation index – Guidelines for the Use of Radio-
J Clin Orthod 2002;36:398–406. in detecting peak pubertal growth in graphs in Clinical Orthodontics, 3rd
15. Ozer T, Kama JD, Ozer SY. A practi- Indonesian and white subjects with edn. London: British Orthodontic
cal method for determining pubertal receiver operating characteristics Society; 2008.
growth spurt. Am J Orthod Dentofa- analysis. Am J Orthod Dentofacial 30. Soegiharto BM, Cunningham SJ,
cial Orthop 2006;130:131 e131–136. Orthop 2008;134:227–37. Moles DR. Skeletal maturation in
16. Wong RW, Alkhal HA, Rabie AB. Use 24. Perinetti G, Franchi L, Castaldo A, Indonesian and white children
of cervical vertebral maturation to Contardo L. Gingival crevicular fluid assessed with hand-wrist and cervi-
determine skeletal age. Am J Orthod protein content and alkaline phos- cal vertebrae methods. Am J Orthod
Dentofacial Orthop 2009;136:484 phatase activity in relation to puber- Dentofacial Orthop 2008;134:217–26.
e481–486; discussion 484–485. tal growth phase. Angle Orthod 31. Franchi L, Baccetti T, McNamara JA.
17. Gwet KL. Inter-Rater Reliability Dis- 2012;82:1047–52. Postpubertal assessment of treat-
cussion Corner. http://agreestat. 25. Masoud M, Masoud I, Kent RL Jr, ment timing for maxillary expansion
com/blog_irr/sample_size_determi- Gowharji N, Cohen LE. Assessing and protraction therapy followed by
nation.html. Last accessed August 9, skeletal maturity by using blood fixed appliances. Am J Orthod
2012; 2010. spot insulin-like growth factor I Dentofacial Orthop 2004;126:555–68.

Orthod Craniofac Res 2014;17:270–279 | 279