Bone-specific median age of hand-wrist maturation from Sudan

Abstract Background Maturation of bones in the hand–wrist region varies among individuals of the same age and among world groups. Although some studies from Africa report differences to other ethnic groups, the lack of detailed bone-specific maturity data prevents meaningful comparisons. Aim The aim of this study was to describe bone-specific maturity for developing hand–wrist bones in individuals in Khartoum, Sudan. Subjects and methods The sample was selected from healthy patients attending a dental hospital in Khartoum with known age and ancestry (males = 280, females = 330; aged between 3 and 25 years). Bones were assessed from radiographs of the left hand and wrist after the Greulich and Pyle Atlas (1959). Median ages of attainment for bone stages were calculated using probit analysis for each stage in males and females separately. Results Maturity data for stages of the phalanges, metacarpals, carpals and radius and ulna in males and females are presented. Median ages in females were earlier compared to males for all stages. These results are largely earlier than previously published findings or where these could be calculated. Conclusion These results of individual maturity stages of phalanges, metacarpals, carpals and the distal epiphyses of the radius and ulna are useful to assess maturity in growing individuals from Sudan.


Introduction
Skeletal maturity is a multi-factorial process that defines the attainment of specific biological markers in individuals and correlates well with somatic growth.Skeletal maturity is used to assess child development and can predict biological age in forensic science.Skeletal maturity is thought by some to be similar in human populations of different origin, in the absence of poor nutrition, history of chronic diseases or infections (Hochberg 2016;Dahlberg et al. 2019).In contrast, a systematic review maintains that ethnic differences in maturity exist, particularly in populations with African origin (Alshamrani et al. 2019).Further studies are required to identify and understand these differences to compare reference data from world populations (Alshamrani et al. 2019).Skeletal maturity of the hand and wrist relies on the presence of particular centres of bone formation as well as their dimension and structure.Most studies that investigate skeletal maturity rely on the use of composite assessment of hand and wrist bones.The most widely used atlas is by Greulich and Pyle (GP), while bone scoring methods include those by Tanner Whitehouse (TW3) amongst others (Greulich and Pyle 1959;Tanner et al. 2001).The GP atlas accounts for the majority of published literature used to assess skeletal maturity around the world.The GP atlas summarised the timeline of the maturity of the entire hand-wrist of medium-high social class individuals of North European origin, previously referred to as Caucasian, in a series of representative median radiographs per age category from birth to 19 years from 1931 to 1942.
Different methodological approaches hamper our understanding of comparative bone maturity.More than 1 year of bone age (i.e.bone age 7 years to bone age 8 years) may not necessarily equal 1 year of chronological age.Interpreting differences between populations for hand-wrist maturity are complicated by several sources of variability.Hand and wrist maturity indicators from different regions within the hand appear at different ages.Variation of up to 18 months for the timing of onset and completion of some stages for hand-wrist bones between individuals has been reported (Greulich and Pyle 1959;Garn et al. 1961).In addition, the duration of bone maturity stages, particularly of the hand bones, can differ considerably between individuals (Garn et al. 1961).Bone age can differ to chronological age by as much as 2 years (Alshamrani et al. 2019;Dahlberg et al. 2019).Different approaches such as the use of a composite atlas (Gruelich and Pyle) versus a weighted bone analysis (TW3) are fundamentally different and are not directly comparable.The Greulich and Pyle atlas is a series of radiographs for each age group representing the median child for each 'bone age'.A given radiograph as a whole is assessed visually by considering the relative maturity of anatomical regions of interest in the hand.Bone age is assigned using the timeline of overall maturity of the entire hand.This approach has been suggested to be used with care as it is a visual interpretation of many bones (Pyle et al. 1961).It is not unusual to observe some bones differ in maturity at the assigned age.Tanner Whitehouse (TW3) represents a standardised assessment of weighted scores of developing individual bones to express maturity as a score equivalent to percentage mature.This is a more detailed bone-specific approach, where individual bones are scored and combined to indicate bone age.TW3 overcomes the challenge of the high variability between different bones (Cole et al. 2015).
The universal applicability of the GP atlas has been questioned particularly to populations from Africa (Alshamrani et al. 2019).This conclusion is largely based on males from Morocco and African-American males and females and these findings may not hold true for East Africans.Differences of up to 2 years in the timing of hand-wrist maturity have been reported from Malawi, Benin, South Africa, Kenya, Ghana, Sudan and Senegal, compared to populations, referred to as Caucasians, using composite and scoring methods, e.g.GP atlas and TW3 (Greulich and Pyle 1959;Tanner et al. 2001).All male populations showed a delay in hand-wrist maturity relative to the reference used (MacKay 1952;Masse and Hunt 1963;Nakamura et al. 1994;Lewis et al. 2002;Agossou-Voyeme et al. 2005;Dembetembe and Morris 2012;Elamin et al. 2017).Not all female populations showed delay (MacKay 1952;Masse and Hunt 1963;Nakamura et al. 1994;Lewis et al. 2002;Agossou-Voyeme et al. 2005), one study reported no difference (Cole et al. 2015), while an inconsistent delay in young females has also been reported (Elamin et al. 2017).Delay did not seem to persist post-puberty (Nakamura et al. 1994;Elamin et al. 2017).Most of these studies report data that are not bone-specific and, therefore, not readily comparable because their results are summative from multiple bones considered together in either GP or TW methods without details of each particular bone.
Two studies present bone maturity data from East Africa (MacKay 1952; Lewis et al. 2002), however, both include individuals with evidence of under-nutrition.Mackay (1952) reported bone and sex-specific maturity for hand and wrist bones compared to a Caucasian group from the United States.Lewis et al. (2002) reported skeletal delay using the GP atlas method.These findings are insufficient to conclude if East Africans differ to other populations.
Bone maturity data from individuals from East Africa is essential because of the recent influx in human migration pattern into Europe and North America (UNHCR 2019).Maturity of young adults is primarily assessed using references based on data from above average socio-economic background Northern Europeans (Greulich and Pyle 1959).A previous study from Khartoum showed that the GP atlas was not applicable to this group (Elamin et al. 2017).The aim of this study was to describe the timing of individual bones in a group of Sudanese individuals that may be used for future comparison with other populations.

Subjects and methods
This study was conducted at the dental hospital of the Khartoum Centre for Research and Medical Training in Sudan.The sample consisted of 609 healthy Sudanese subjects (280 males and 330 females) from 3-25 years (Table 1) drawn from an earlier sample (Elamin et al. 2017).Older ages were included to ensure sufficient individuals who were fully mature.Exclusions were individuals receiving specialist medical care.On attendance, the medical history and date of birth were recorded or verified from accompanying parents in the case of minors.Ethical approval was granted by the Ethics committee at the Centre before the study (1 November 2010).Verbal and written consent was obtained from study subjects and parents of minors.A digital radiograph of the left hand was obtained (CS8100, Carestream; USA) and viewed without image adjustment.Every visible bone was assigned a maturational stage from the GP atlas by the second author (HYHM) using the line drawings and descriptions of individual maturity stages of the distal epiphyses of the radius, ulna, carpals, metacarpals, proximal, middle and distal phalanges in the GP atlas (Greulich and Pyle 1959, pp. 152-190).Intra-examiner agreement was assessed from 20 radiographs and Kappa was excellent at 0.87 (Landis and Koch 1977).The median age of attainment was calculated using probit regression analysis for each stage of each individual bone where sufficient data allowed.

Results
The results of this descriptive study are detailed in Tables 2-5 with median ages for individual stages for the developing bones in males and females across 3-25 years of age.Results are tabulated according to hand region.Table 2 details GP stages of the phalanges.Table 3 shows the results of metacarpal stages and Table 4 stages of the carpal bones.Median ages of the radius and ulna are shown in Table 5. Raw data for the radius are given in Table 6 and the percentage of males and females with radius stage 13 (mature) for each year of age are illustrated in Figure 1.There are more In all instances, median ages were earlier in females than in males.This difference could be as large as 2 years.In a number of instances, carpal bones were late maturing and it was not possible to calculate ages of full maturity.

Discussion
Our study details the timing of the maturity of individual bones in the hand and wrist region from individuals in Sudan for the first time.There is a lack of bone-specific data from this region.Only a few previous studies have documented information about the timing of appearance of hand-wrist bones in Africa.The first was a study of skeletal maturation in the hand from birth to 18 years of age of the Wadiga tribe in East Africa in southern Kenya/northern Tanzania (MacKay 1952).The percentage of boys and girls showing the start and complete fusion of individual bones are tabulated as a percentage for each chronological year of age.These results were compared to children in Chicago and showed that the children from East Africa were delayed in their start/completion of fusion of bones.A major contribution from this study is the illustration of median hand-wrist radiographs for each year of age from age 6 months to 17 years and 6 months in males and females.The second study was from Senegal, West Africa (Masse and Hunt 1963), describing the skeletal maturation of hand and wrist from birth to 15 years of age.Median ages of GP stages of individual bones for girls are tabulated.Girls from Dakar were earlier in early childhood compared to the reference girls.Ages were similar between the groups at the age of 18 months and, thereafter, girls from Dakar were considerably delayed, although only some GP stages were compared.Both these studies present more data for maturation in infants and young children than older children.
Our results are generally somewhat earlier than previously published results of individual bone maturity.Median ages are similar to five mature bone stages for boys and girls reported by Pyle et al. (1961).Median ages for girls in our study are earlier than girls from Dakar (Masse and Hunt 1963).Our results for boys compared to GP stages of individual bones (Greulich and Pyle 1959, pp. 152-190) showed that 91 out of 116 stages (78%) were earlier, 24 similar in age and one stage was later than the reference.For girls, the timing of stages compared to GP showed that 50 out of 62 stages (81%) were earlier, with 12 being similar to the reference.Boeyer et al. (2018) notes a small secular trend of advancement in the timing of epiphyseal union.Our results are largely earlier than the ages they report for complete fusion of some bones (6/9 stages in boys and 6/7 in girls).
Our results of median age of complete epiphyseal union of the radius and ulna are considerably earlier than other results that we have calculated from published summary data.Several studies tabulate the number of individuals per stage/chronological year of age for the distal epiphysis of radius and ulna allowing median age to be calculated.A study from India found no difference in stages of the radius and ulna and tabulated data for these combined (Banerjee and Agarwal 1998).Median age was calculated as 17.3 in males and 15.7 in females.Median ages in individuals from Kashmir, Northern India, were calculated as 16.8 and 15.8 for the radius and 16.5 and 15.5 for the ulna for males and females, respectively (Hassan et al. 2016).Another study from western India combines data for left and right side radiographs and median age was calculated as 18.8 and 18.3 for the radius and 18.5 and 17.7 for the ulna for males and females, respectively (Patel et al. 2011).Another western Indian study (Rajdev et al. 2014) provides data and median ages for the radius were calculated as 18.6 and 17.8 and for the ulna 18.1 and 16.7 for males and females, respectively (small N for females).In a study of girls only from north western India, median age was calculated as 17.6 for the radius and 17.2 for the ulna (Sahni et al. 1995).A study from Jordan tabulates such data; however, the numbers for each year of age are small (Al-Qtaitat et al. 2016).The median age calculated using probit for males was 18.1 for the radius and 18.6 for the ulna.One study from Saudi Arabia reports on the hand (metacarpals and phalanges) with tabulated data/ stage of the number of individuals per year of age, however the numbers for each year are very small (Aljuaid and El-Ghamry 2018).A study from Nigeria tabulates data but the sample size per year is very small (Ebeye et al. 2021), although the two 18-year old males and all 12 19-year old males showed complete fusion of both radius and ulna.Another study from Nigeria gives percentages per year with incomplete and complete union of the radius and ulna with no sample size details (Ominde et al. 2018).Other studies The evidence that Africans differ to North American or British populations is based on studies using atlas or composite methods of bone maturity that summarise the maturity of the hand-wrist as a whole, with no data for individual bones.The reported delay in bone age in males from Africa using atlas-based or composite methods (MacKay 1952; Masse and Hunt 1963;Nakamura et al. 1994;Lewis et al. 2002;Agossou-Voyeme et al. 2005;Dembetembe and Morris 2012;Elamin et al. 2017) is in contrast to our findings of considerably earlier median ages (both males and females) of individual bone stages compared to the GP reference.We compared our results to the reported age or age range for individual GP bone stages (Greulich and Pyle 1959, pp. 152-190).These ages are not calculated for that bone-specific stage but are the median radiographic reference and this comparison may not be valid.Consecutive stages of individual bones rely on sometimes small, morphologic changes of shape and dimension that are subjective.Clearly this finding of earlier median ages and previously delayed overall bone age delay warrants further investigation.
The main strength of our study is that it presents ages for individual maturity stages of phalanges, metacarpals, carpals and the distal epiphyses of the radius and ulna.The age range, including older ages, is also a strength, as many studies fail to include sufficient older individuals.By including older individuals in our sample, the cumulative statistical approach allows us to assess full maturity of the radius (including the age when all individuals reach maturity).The lack of such data has been a limitation in the application of age estimation in young adults in a forensic context.Recent migrations into Europe highlight the lack of contemporaneous maturity data.Our sample is from Africa, an area not well represented in the literature.The only previous study documenting bone-specific data from East Africa lists frequency tables and median radiographs from a group showing signs of malnutrition (MacKay 1952).Lewis et al. (2002) also showed a delay in this group compared to the GP reference, noting poor living conditions.
This study has several limitations.The GP atlas is based on maturing individuals from Ohio in optimal conditions.Our study describes skeletal maturity in prevailing conditions in East Africa, and a representative sample will inevitably reflect environment stress.Our study was a cross-sectional convenience sample of healthy individuals and young adults, but our sample lacked young children and some age categories, particularly girls, are not well represented.The culture is traditionally tribal with more than a hundred languages and limited inter-marriage culture between tribes.The tribes reside north of the fifth cataract and have a similar Nubian Arab heritage.Sudan has a high-stress environment and continues to be affected by civil war, endemic tropical disease and poor living conditions.However, the individuals from this study were considered medically fit and had middle-tohigh socioeconomic status as they attended privately-funded dental and orthodontic treatment.The sample includes individuals and adults of 3-25-years of age to include older ages when all hand-wrist bones are mature, although a number of older individuals showed delayed carpal bones.Including older individuals in this study addresses the criticism that many previous studies truncated the maximum age of samples at 18, irrespective of whether all individuals are fully mature.Our results are tabulated in detail and are not readily usable in everyday applications in this format.Future work will summarise and illustrate these results in the form of an atlas.
Our conclusion, based on the evidence of these results, is that more data are needed from other regions and such bone-specific data should be tabulated against age categories (such as Table 6).

Figure 1 .
Figure 1.Percentage of males and females with radius stage 13 (mature) for each year of age.Solid line males, dashed line females, vertical line at age 18 years.

Table 1 .
Age and sex distribution of sample.
Age group 3.5 indicates age 3.00-3.99years,etc.calculate the cumulative frequency and large sample numbers are necessary in each year of age to fully describe the timing of all bone stages.This is evident in Table6, where raw data are tabulated.Stages 10 and 11 were only observed in two males.

Table 2 .
Bone-specific median ages for phalange bones in years calculated using probit analysis for 280 males and 330 females from North Sudan.

Table 3 .
Bone-specific median ages for maturing metacarpal bones in years calculated using probit analysis for 280 males and 330 females from North Sudan.

Table 4 .
Bone-specific median ages of carpal bones in years calculated using probit analysis for 280 males and 330 females from North Sudan.

Table 5 .
Bone-specific median ages of the radius and ulna in years calculated using probit analysis for 280 males and 330 females from North Sudan.

Table 6 .
Number of individuals per stage of the radius observed in males and females per chronological age group.
(Schmidt et al. 2008;Baumann et al. 2009ative median age(Schmidt et al. 2008;Baumann et al. 2009) and it is inappropriate to compare with median age calculated from cumulative frequency distribution methods such as probit or transition age.