Introduction

Fingerprint is one of the widely used biometrics for the purpose of human identification. Its uniqueness, consistency, and inexpensiveness give room for its application in the distinction between individuals at different stages of human life (Zugibe and Costello, 1986; Yager and Amin, 2004; Gutierrez et al., 2007; Kahana et al., 2001; Franca, 2011; Vanrell, 2012). The fingerprint ridge configuration formed by epidermal ridges during 12th–19th gestational weeks remains fixed and permanent throughout the life of an individual (Mulvihill and Smith, 1969; Babler, 1979). The epidermal ridges are explained at different levels; these include pattern configuration, minutiae, ridge characteristic such as ridge flow, shape, contour, etc. (Hong et al., 1998; Jain et al., 2007). The epidermal ridges are often associated with surface wrinkling on gross examination of fingertip called fingerprint white lines.

Fingerprint white lines (FWL) correspond to areas of depression on fingerprints that appear white on the fingerprint images and skin deformation on fingertip that may be linked to loss of the elements of skin tissues (Cummins and Midlo, 1943; Ashbaugh, 1999; D’Adamo, 2010; Maceo, 2011). Previous studies suggested that with aging, there is decrease in the strata of the epithelium and an inter digitations between epidermis and dermis of the skin (Makrantonaki and Zouboulis, 2007; Waaijer et al., 2012; Farage, 2013) and reduction of elastin synthesis (Farage, 2013). This leads to decrease in the elasticity and support of the tissue, increase in the brittleness, and impairment of nuclear and mitochondrial DNA repair mechanisms. Hence, the chance for appearance of lesions and spots was exemplified by FWL (Makrantonaki and Zouboulis, 2007). This mechanism is supported by a documented increase in FWL in elderly (Vieira Silva et al., 2016) and increase in frequency of FWL with age or when there is alteration in subcutaneous body fat (Cummins and Midlo, 1943; Ashbaugh, 1999).

It was hypothesized that the size of a fetal fingertip (compared with its neighbor on the same hand) might be influenced by factors that stimulate or inhibit growth along the developmental axis extending from the brain to the lower limbs (Kahn et al., 2001). It is well documented that the growth along developmental axis is sexually dimorphic, with upper to lower body segments proportional difference been higher in males than in females (Bailey and Katch, 1981). However, the potential roles of differences in adjacent FWLC in sex inference are not well elucidated in the literature especially among Hausa population. Only few studies reported the potential role of FWL in sex determination in the other populations (Badawi et al., 2006; Taduran et al., 2016) including the Hausa population (Adamu et al., 2019). Since sex determination is one of the leading criteria in identification and verification of an individual among all the key parameters of identification (Ubelaker, 1996), and determination of sex of human remains has been proven to help forensic experts during analyses of dismembered and fragmentary remains and also in criminal investigation by narrowing the pool of potential suspect matches (Kanchan et al., 2008; Dey and Kapoor, 2015). It is, therefore, plausible to look at the utility of FWL in sex determination. The present study was set to achieve the following objectives: (i) to determine the sexual dimorphism in adj. DFWLCD, (ii) to predict sex from the adj. DFWLCD, and (iii) to determine the best adj. DFWLCD that discriminate the sex, among Hausa population of Kano state, Nigeria.

Materials and methods

Study location, design, and participants

The study was carried out at Bayero University Kano (a Federal university) and Maitama Sule University, Kano (a state University). This was a cross sectional study conducted among students of the two selected tertiary institutions in Kano metropolis, Nigeria. We randomly selected sample of 150 males and 150 females to participate in the study based on the study selection criteria. The participants were within the age range of 18–33 years (mean age of 21.94 ± 2.31 and 20.13 ± 2.33 years for male and female respectively). We excluded any participant with physical deformity in the tip of their digit and belonged to other ethnic group. We adopted self-declaration of the ethnicity methods, and only those that belong to Hausa ethnic group up to the level of grand parentage were considered. We obtained an informed consent from the participants before commencement of the study. The study allowed decline of consent by participants at any stage of the data collection (although none of the participants declined his/her consent). We used simple proforma for collection of bio-data (sex, age, and ethnicity) of the participants. The study was conducted following the ethical guidelines of Helsinki Declaration. The protocols involved in the study were approved by the Department of Anatomy, Faculty of Basic Medical Sciences, College of Health Science, Bayero University Kano.

Fingerprints capturing and white line counts

A live scanner (Digita Persona, China) was used to capture the fingerprint of all the ten digits according to methods described by Adamu et al. (2018). We defined the white line as the skin folds in the friction fingerprint epidermal ridges that appear as white lines in the fingerprint images (Fig. 1). The FWLC for each digit was considered as the number of the observed white lines per unit plain fingerprints (Taduran et al., 2016; Adamu et al., 2019). The reliability and intra observer error in the FWLC were reported elsewhere (Adamu et al., 2019).

Fig. 1
figure 1

Techniques for fingerprints white line counts

Full size image

Determination of adjacent digit FWLC differences

It was suggested that each tip of digit is related neurologically to a spinal-cord segment in a range that includes the sixth (C6) through the eighth cervical (C8) levels. The first digit (thumb) is linked to the cephalad (upper) side of C6, and the fifth digit is linked to the caudal (lower) side of C8 (Heimer, 1994). We determined the adj. DFWLCD as per Kahn et al. (2001) methods of determination of adjacent fingerprint ridge count difference. The adj. DFWLCD was therefore determined as the differences in FWLC from the caudal FWLC value subtracted from the cephalad FWLC value (e.g., dR12 is right first digit (thumb) FWLC minus right second digit (index) FWLC, that is the difference in FWLC between thumb and index finger).

Statistical analyses

The data were expressed using descriptive statistics (mean ± SD, minimum, and maximum), median (interquartile range (25th and 75th percentiles)). The normality test (Shapiro Wilk test, P < 0.05) was also carried out. Mann-Whitney and t test were used for comparison of variable between sexes. Binary logistic regression analyses were used to generate a model for sex prediction using adj. DFWLCD. Stepwise multiple (forward conditional) binary logistic regression analysis was also conducted to determine the adj. DFWLCD that predict sex. The analyses were carried out using SPSS version 20 (IBM Corporation, for Windows) and 5% (P < 0.05) level of error was considered for any statistical inference.

Results

Table 1 shows the descriptive statistics of the FWLC of the right and left hands digits. The mean FWLC was higher in the females in all the ten digits. Absence of FWLC was observed in both sexes in all the digits as the minimum counts. However, the maximum FWLC was in favor of females, with higher count across the ten digits.

Table 1 Descriptive statistics of the FWLC of the right and left hands digits
Full size table

From Table 2, the FWLC on the right digits was found to be higher on the first digit compared with other adjacent digits (second, third, fourth, and fifth) in both sexes except for dR14 in females where fourth digit had higher FWLC compared with the first digit, as indicated by negative mean value (− 0.23 ± 2.18). The second and third digits were observed to have lower FWLC compared with other adjacent digits in both males and females, except for dR35 in males where third digit had higher FWLC compared with the fifth digit, as indicated by positive mean value (0.07 ± 1.02). The mean dR45 was positive in both sexes in the right hand. Similar trend was observed in the left hand except in dL25 in males where the FWLC of the second digit was higher than that of fifth digit, as indicated by positive mean value (0.19 ± 1.27) and dL15 in females where the FWLC of the second digit was lower than the that of fifth digit, as indicated by negative mean value (− 0.05 ± 2.41). There were significantly larger adj. DFWLCD in males compared with females in most of the digit combination. However, few digits indicated higher adjacent digit FWLC difference adj. DFWLCD in females, but none was significant.

Table 2 Descriptive statistics adj. DFWLCD and sexual dimorphism in cephalic to caudal growth inhibition in the right and left sides of the hand
Full size table

Table 3 shows sexual dimorphism in FWLC adjacent digit asymmetry of the right and left digits. A significant sexual dimorphism was observed in all the adjacent FWLC differences involving ring digit in both right (dR14, dR24, and dR34) and left (dL14, dL24, and dL34) digits except for dR45. Similarly, dR25 and dL25 exhibited significant sexual dimorphism. The magnitude of the sexual dimorphism was more pronounced in dR24 (Z = − 3.69, P < 0.001) and dL24 (Z = − 4.11, P < 0.001), more in the left that in the right.

Table 3 Sexual dimorphism in FWLC adjacent digit asymmetry of the right and left digits
Full size table

A significant sex prediction potential of adj. DFWLCD was observed in dR14, dR15, dR23, dR24, dR25, and dR34. Similar trend was observed in the left digits except for dL23 and addition of dL35. The best discrimination was observed in adj. DFWLCD of second and fourth digits in both right and left digits (dR24 and dL24). This was further supported by stepwise (forward conditional) multiple binary logistic regression analyses. The variance of sex explained by adjacent digit FWLC difference was higher in dL24 (7.4 to 9.9%) compared dR24 (5.6 to 7.4%) as shown in Table 4.

Table 4 Sex prediction using FWLC adjacent difference of the right and left digits among Hausa population
Full size table

Discussion

Identification of human remains is an essential element of any medico-legal investigation, and it has been a demanding task for forensic experts and physical anthropologists across the globe (Kanchan et al., 2008). This has led to continuous need to explore different body parts and techniques to determine key identity parameters often collectively described as “the Big Four”; race, sex, age, and stature (Scheuer, 2002; Ahmed and Omer, 2015). Therefore, the goal of the present study was to investigate and provide insight in to the role of adj. DFWLCD as a probable additional forensic tool that might compliment the sex component of “the Big Four.”

The difference in adj. DFWLCD was significantly more pronounced in males compared with females as an indication of larger cephalocaudal growth inhibition in males. This is supported by a previous study, where large fingerprint ridge-count difference between fingers on the same hand (a decline in the cephalocaudal direction) was used to reflect conditions associated with relative inhibition of caudal growth (Kahn et al., 2001). The concept of a more pronounced caudal growth inhibition in males may be explained by the fact that sexual dimorphism in the circumference of the upper limb (averaged 17.9%) was higher compared to that in the lower limb (4.6%), and this exhibited fairly consistent proximal to distal dimorphism in most of the measured body parameters (Bailey and Katch, 1981). It was also suggested that growth trends in the limbs are essentially linear through late adolescence but completed earlier in females (Hiernaux, 1968). The adult leg would therefore be less dimorphic than the adult arm (Bailey and Katch, 1981). Also, according to Kahn et al. (2001), developing embryos with higher adjacent digit ridge count difference might accumulate relatively less tissue in the lower body. This might express reduced lower-extremity muscle mass and tissue distribution compared with the upper body. From the foregoing explanations, it is clear that any measurable parameter that serves as an indicator of caudal inhibition, including adj. DFWLCD, will be of interest to forensic experts especially with regard to sex determination.

The significant sexual dimorphism observed in the adj. DFWLCD in the present study; reaffirms the potential of the FWLC in sex prediction as reported in previous studies (Badawi et al., 2006; Taduran et al., 2016; Adamu et al., 2019). The existence of sexual dimorphism in the adj. DFWLCD may be superior to the absolute FWLC in sex prediction especially among the elderly population, since, it was demonstrated that the FWL increases from adult to elderly life cycle. This is linked to the loss of resilience of the skin, which is the property of returning of the skin to its original shape after being subjected to elastic deformation (Vieira Silva et al., 2016). Frequency of FWL was also reported to increase with age or when there is alteration in subcutaneous body fat (Cummins and Midlo, 1943; Ashbaugh, 1999). In addition to FWL changes, the epidermal ridge thickness and inter papillary space were also reported to increase with age as a continuum of increase in body size (Cummins et al., 1994; David, 1981; Babler, 1991; Gutierrez-Redomero et al., 2011; Soanboon et al., 2016; Sánchez-Andrés et al., 2018). But, the adjacent finger ridge count difference (dR45) was suggested to be consistent with aging (Kahn et al., 2001). We therefore extended the same observation, of consistency of finger ridge count difference with aging, to adj. DFWLCD. This may be explained by the fact that, similar to second-to-fourth digit ratio of the hand (2D:4D), the indices do not change with aging and growth in body parts (Manning et al., 2004; Aboul-Hagag et al., 2011). Hence, it was concluded that sexing by indices is more reliable than absolute measurement since the relative growth of body parts are proportional to each other (Gutnik et al., 2015). This may lead to the assumption that adj. DFWLCD might not be altered by age or change in body condition similar to adjacent finger ridge count difference and digit ratios.

The present quantification method (adj. DFWLCD) of FWLC may not only be superior to absolute quantification of FWLC but also to bilateral asymmetry in FWLC. It was documented that the occurrence of bilateral asymmetry depends on the extent exposure to stress and strain on the dominant side (Gutnik et al., 2015) This implies that the greater the use of the preferred side, the more the hypertrophy and development of the muscles on that side of the limb with consequential increase in asymmetry (Plochocki, 2004; Blackburn, 2011; Krishan, 2011; Krzykala and Leszczynski, 2015). Although, the influence of testosterone is more pronounced on the right hand even among the left handed individuals (Hampson and Sankar, 2012), and not more than 10% of every population are left handed, (Hardyck and Petrinovich, 1977) and 5% among African ages 18 to 33 years (De Agostini et al., 1997). For adj. DFWLCD, the differential influences of hand preference are minimal or absent considering the fact that both digits belong to the same body side.

Interestingly, in both right and left hands, the adj. DFWLCD in second and fourth digits (dR24 and dL24) was the best predictor of sex. This is similar to what was reported for digit ratio, where the 2D:4D exhibited the highest magnitude of sex discrimination especially in the right hand (Kyriakidis and Papaioannidou, 2008; Agnihotri et al., 2015). The 2D:4D is negatively correlated with prenatal testosterone and positively correlated with estrogen (Manning et al., 1998). This relationship between the digit ratio and sex hormones may be extended to other features associated with the digits (second and fourth digits) in a direct or an inverse manner. For the first time, we hypothesized that the dR24 and dL24 of FWLC correlated positively to prenatal testosterone and negatively correlated to estrogen. This may also be extended to other associated features like fingerprint ridge count which has been demonstrated to be an index of caudal growth inhibition (Kahn et al., 2001). Indeed, we also suggested inverse relationship between the adjacent digits associated feature differences and digit ratios, with higher value indicating “masculine feature” and lower value as “feminine feature” and also, more expression of adj. DFWLCD in the left (dL24) than the right (dL24). This is somewhat a paradox when compared with digit ratio, where high and low 2D:4D has been suggested to be a “feminine and masculine features” respectively (Manning et al., 1998). And also 2D:4D is more sexually dimorphic on the right compared with the left in human hands (Manning et al., 1998; Manning, 2002), the feet of mice (Brown et al., 2002a; Brown et al., 2002), and zebra fishes (Burley and Foster, 2004). Though, further studies are needed to evaluate the direct relationship of adjacent digit associated feature differences with prenatal testosterone and estrogen.

In the present study, we highlighted the significance of adj. DFWLCD with special interest on FWLC as an index for sex discrimination. The pattern of sexual dimorphism exhibited is similar to well established sexually dimorphic digit ratio. This particular finding is a pointer to the need to investigate whether sexual dimorphism in adj. DFWLCD as observed in the present study is a universal phenomenon equally present in populations of different races and ethnicity. Considering the sex prediction potential exhibited by adj. DFWLCD, this study would be useful in any forensic case work since it offers adj. DFWLCD as a new biological characteristic that could compliment other known sex discriminating tools in forensic science. The study may also by extension serve as a source of reference data for forensic anthropologists in the study population. This can be applied in the analyses of cases in situations like mass disasters, acts of violence, bombing, and traffic accidents, which may be associated with dismembered human remains and peripheral parts of the body. It can be applied in assault cases where the victim’s body is dismembered to conceal identity. Proper identification of dismembered, mutilated, and fragmentary remains using adj. DFWLCD may be imperative.

Conclusion

We concluded that adj. DFWLCD exhibited a sexual dimorphism phenomenon that can be used in sex prediction. The best prediction potential was found to be dR24 and dL24 for right and left hands respectively among Hausa population of Kano state, Nigeria. The adj. DFWLCD is an important index for evaluation of caudal growth inhibition along the body axis, a well-established sexually dimorphic phenomenon in human population.