Sie sind auf Seite 1von 28

Research Article

Bacterial and Parasitic Assessment from Fingernails in Debre Markos, Northwest Ethiopia
Abeba Mengist, Yibeltal Aschale, and Alemayehu Reta

Department of Medical Laboratory Science, College of Medical and Health Sciences, Debre Markos University, Debre
Markos, Ethiopia

Correspondence should be addressed to Abeba Mengist;

Received 7 July 2018; Revised 8 September 2018; Accepted 27 September 2018; Published 18 October 2018

Guest Editor: Nicola Serra

Copyright © 2018 Abeba Mengist et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly


Background. Food handlers with untrimmed finger nails could contribute or serve as a vehicle for the transmission of food
poisoning pathogens. Objectives. This study was conducted to determine the prevalence of bacteria and intestinal parasites
among food handlers and antibiotic susceptibility profile of the isolated bacteria in Debre Markos University,
Ethiopia. Materials and Methods. This laboratory-based cross-sectional study involved 220 food handlers working in food
service establishments in Debre Markos University between 1st January 2015 to 31th June 2016. Subjects’ finger nail
specimens of both hands were examined microscopically for intestinal parasites. For bacterial isolation, samples were
cultured and bacterial species were identified following standard laboratory procedures. Antimicrobial susceptibility test was
performed for all bacterial isolates by using Kirby-Bauer disk diffusion method. Results. Of the total 220 subjects examined,
29.5% showed positive culture for different bacterial species from their fingernail contents. Coagulase-negative
Staphylococcus was the predominant bacteria species (12.3%) followed by Staphylococcus aureus (5%), E.
coli (2.7%), Klebsiella species (2.7%), Enterococcus species (1.8%), Pseudomonas aeruginosa (1.8%), Proteus
species (1.4%), Citrobacter species (1.4%), and Serratia species (0.9%). None of the food handlers showed positive culture
for Shigella and salmonella and parasites in respect of their finger nail specimens. Isolation of bacteria in finger nail has
significant association with finger nail status () and inverse relation with service years (). All Staphylococcus aureus and
coagulase-negative Staphylococcus species isolates were uniformly susceptible to vancomycin. Only one (9.1%)
of Staphylococcus aureus isolates was resistant for methicillin. Conclusion. To prevent the food poisoning pathogens,
implementation and adherence to infection are the key practices, specially food handlers with long finger nail harbor food
debris, microbial contaminations, and allergens.

1. Background

Food-borne disease is a public health problem in developed and developing countries due to poor food handling and
sanitation habit, inadequate food safety programs, lack of clean water supply, poverty, and lack of knowledge of food
handlers. According to the World Health Organization (WHO) statement, most of the populations suffer from food-borne
diseases every year in both developed and developing countries .The spread of food-borne disease through food handlers
is a common and persistent problem worldwide .Infected food handlers with poor hygiene practice working in food service
establishments are potential sources and transmitters of the disease due to pathogenic organisms like infection with various
intestinal helminths, protozoal, and enteropathogenic bacteria .They can transmit both enteric and nonenteric bacterial and
parasitic infections through the food that they handled .
Microorganisms such as bacteria, parasites, and viruses are the common agents for food contamination. Vibrio cholera,
Campylobacter jejuni, Enterotoxigenic Escherichia coli, Salmonella typhi, Shigella species, and Polio are the most common
food-borne disease causing organisms in developing countries .Protozoan and helminthic parasites such as Giardia
lamblia, Entamoeba histolytica, Cryptosporidium species, Ascaris lumbricoides, and Enterobius vermicularis are also
important agents of food-borne disease. These infections in food handlers pose a significant threat to food consumers
.Transmission of finger nail bacteria occurs through food, water, nails, and fingers contaminated with feces demonstrating
the role of fecal-oral person-to-person transmission .Food handlers who harbor and excrete bacteria may contaminate foods
from their feces via their fingers and then to food preparation and servicing and lastly infect healthy individuals .The area
under the fingernail spreads pathogenic microorganisms via cross contamination, and it is challenging to clean when
compared with other parts of the hand .Food storage systems such as temperature and time, food preparation, handling,
servicing practices, and food handlers’ knowledge and skill are some of the factors that affect the safety of food directly or
indirectly .Biofilm formation is an important factor in persistence of microorganisms on the surface. Cells in a biofilm are
embedded in an extracellular polymeric matrix constituent, proving resistant to conventional therapeutic doses of
antimicrobial agents and clearance by the host response. Biofilm formation proceeds via initial adhesion to the surface and
subsequent aggregation into multicellular structures. Thus, the development of a biofilm requires adhesive forces for the
colonization of surfaces and cell interaction. Specifically, S. epidermidisis, one of the major biofilm-producing bacteria,
works by attaching itself to several surfaces .Various measures have been implemented to reduce incidence of food-borne
diseases both in developed and developing countries. However, there has been increased occurrence of emerging and
reemerging food-borne diseases. Among the factors responsible for this is the resistance of food-borne pathogens to
antibiotics. Humans are exposed to resistant bacteria through sources such as food products, environment, and food
handlers. Among the factors responsible for this occurrence and prevalence are poor food-production processes,
inadequate food storage infrastructure, unhygienic food handling, limited resources, and poorly enforced regulatory
standards .

Therefore, appropriate screening method is useful to detect bacterial and parasitic infections among food handler’s finger
nails, thus preventing probable illness and protecting the health of the consumers. Thus, this study was carried out to
determine the prevalence and susceptibility pattern of finger nail bacteria and parasites among food handlers in Debre
Markos University food service establishments to address appropriate recommendations for the enhancement of good food
safety and sanitary conditions within food establishments in the University.

2. Materials and Methods

2.1. Study Design, Area, Period, and Participants

The present laboratory-based cross-sectional survey included 220 food handlers working in food establishments of Debre
Markos University during the period from January 2015 to June 2016. The University is found in northwestern part of
Ethiopia at Debre Markos town. The town is located 300 km northwest of Addis Ababa.
2.2. Data Collection

Data were collected by the data collectors after obtaining written informed consent using a well-structured questionnaire
designed to obtain sociodemographic data and other relevant data related to food handlers’ service year, status of medical
screening, status of certification, education, and hand-hygiene practices from participants following their written informed
consent and the ethical approval of the study from Debre Markos University ethics review board.
2.3. Sample Collection and Transport
Swab samples under the finger nails from both hands of each subject were collected using sterile-moistened cotton-tipped
swabs and placed into a sterile test tube. Until inoculated on to respective cultured media, the samples were kept with
normal saline in a test tube for not more than 5 minutes .
2.4. Culture and Identification
2.4.1. Processing of Fingernail Swabs and Identification of Bacteria and Parasites

A single under finger nail swab obtained from each food handler was cultured immediately on Mannitol salt agar (MSA) for
isolation of S. aureus and Coagulase-Negative Staphylococci (CNS). Finger nail swabs were cultured on to Salmonella-
Shigella agar (Oxoid), MacConkey agar (Difco), and Blood agar (Oxoid) and then incubated at 37°C for 24 hours for
isolation of Gram-negative bacteria. The bacterial colonies grown on the agar media were presumptively identified by
colonial morphology and gram staining and a battery of biochemical tests like reaction on oxidase, catalase, simmon citrate,
indole production, urease, motility, KIA, and gas and hydrogen sulfide (H2S) production .For parasite identification, samples
were examined microscopically following direct wet mount preparations in normal saline and iodine solution .
2.4.2. Antimicrobial Susceptibility Testing

Antimicrobial susceptibility tests were performed on Muller Hinton Agar (Oxoid, Hampshire, UK) by disc diffusion method.
The following antimicrobial agents were used for Gram-positive isolates: methicillin (10 µg), penicillin (10 µg), erythromycin
(15 µg), ampicillin (30 µg), ciprofloxacin (10 µg), tetracycline (30 µg), cotrimoxazole (25 µg), and vancomycin (30 µg). To
characterize Gram-negative isolates, ampicillin (10 µg), tetracycline (30 µg), chloramphenicol (30 µg), gentamicin (10 µg)
and norfloxacillin (10 µg) and cotrimoxazole (25 µg) and ciprofloxacin (10 µg) have been used. The susceptibility profiles
(i.e., resistance and sensitivity) of the isolates were interpreted according to the National Committee for Clinical Laboratory
Standards .
2.4.3. Data Processing and Analysis

All statistical calculations were done using SPSS for windows version 20. Descriptive statistics were computed to determine
the rate of bacteria and other variables. The relationships between the presence of bacteria and various risk factors were
tested using the Chi square test. A value of ≤0.05 was considered indicative of a statistically significant.

3. Result

3.1. Sociodemographic Data

Two hundred twenty food handlers were participated in this study. Among them, 69.1% were females and 30.9% were
males. The age of the study participants ranged from 18 through 43 with a mean age of 25.1 (SD ± 4.1). Regarding their
job, 45.9% had one to two years of work experience and only 30.9% had more than two years of work experience (Table 1).
Table 1: Sociodemographic characteristics of food handlers () working at food service establishments in Debre Markos
University (1st January 2015 to 31st June 2016).
In this study, the majority (97.3%) and only few (17%) of food handlers had a habit of hand washing after toilet and after
touching different body parts, respectively (Table 2).
Table 2: Hygienic practices of food handlers (), in relation to finger nail bacterial isolates, working at food service
establishments in Debre Markos University (1st January 2015 to 31st June 2016).

3.2. Prevalence of Bacteria Isolated from Finger Nail of Food Handlers

The frequency and type of bacteria isolated from fingernail content of the 220 food handlers studied are presented in
Table 3. Bacteria isolated include coagulase-negative Staphylococcus (12.3%), Staphylococcus aureus (5%), Escherichia
coli (2.7%), Klebsiella species (2.7%), Enterococcus species (1.8%), Pseudomonas aeruginosa (1.8%), Proteus species
(1.4%), Citrobacter species (1.4%), and Serratia species (0.9%). While no bacteria were isolated from the finger nail content
of 70.5% of participants. None of the food handlers showed positive culture for Shigella and salmonella in respect of their
finger nail specimens. No more than one enteric bacterium was observed in the subject under study. In addition, no
intestinal parasites were detected from the samples of fingernail contents.
Table 3: Types of bacteria isolated from finger nail content of food handlers () working at food service establishments in
Debre Markos University (1st January 2015 to 31st June 2016).

In this study, different factors were assessed for possible association with finger nail bacterial isolation rate among the study
participants . The number of positive cultures from finger nail contents was higher among female subjects (30.7%) than
those of male subjects (26.9%), but the difference was not statistically significant () .
Table 4: Sociodemographic characteristics of food handlers (), in relation to finger nail bacterial positivity, working at food
service establishments in Debre Markos University (1st January 2015 to 31st June 2016).

The isolation rate of bacteria in finger nail of food handlers was relatively higher 22(43.1%) among food handlers served for
a period of less than one year and lower 16(23.5%)s among those served for a period of greater than 2 years .Therefore,
the inverse relationship between service year and finger nail bacterial isolation rate was statistically significant (). In addition,
food handlers with long finger nails showed more 33(37.1%)s bacterial isolation rate with their finger nails as compared to
those food handlers with short (properly cut) finger nails 32(24.4%) () . However, the other expected risk factors (i.e., age,
educational background, medical check-up, food hygiene training, and hand washing habit) had not been found to be
associated with bacterial fingernail rate .
3.3. Antimicrobial Susceptibility Pattern of Isolated Pathogens

All Staphylococcus aureus and coagulase-negative Staphylococcus species isolates were uniformly susceptible to
vancomycin. Relatively, Staphylococcus aureus showed low resistance to methicillin (9.1%), cirofloxacin (9.1%) and
erythromycin (18.2%), and cotrimoxazole (18.2%); high resistance to penicillin (63.6%) and ampicillin (63.6%) followed by
amoxycillin and tetracycline with 54.5%% and 45.5%, respectively .The susceptibility profile of the Gram-negative isolates is
presented in Table 6.
Table 5: Antimicrobial resistance pattern of S. aureus and CNS isolated from finger nail cultures of food handlers working at
food service establishments in Debre Markos University (1st January 2015 to 31st June 2016).
Table 6: Antimicrobial resistance pattern of different Gram-negative bacterial species isolated from finger nail content of
food handlers working at food service establishments in Debre Markos University (1st January 2015 to 31st June 2016).

4. Discussion

Improper food handlings practices by food handlers may cause food contamination and food-borne diseases, which may
pose a possible risk to community or customers . Therefore, this study was undertaken to assess the prevalence of bacteria
and intestinal parasites among food handlers and antibiotic susceptibility profile of the isolated bacteria in Debre Markos
University, Ethiopia.
In this study of 220 food handlers examined, 29.5% were carriers of enteric bacteria including coagulase-negative
Staphylococcus (12.3%), Staphylococcus aureus (5%), Escherichia coli (2.7%), Klebsiella species
(2.7%), Enterococcus species (1.8%), Pseudomonas aeruginosa (1.8%), Proteus species (1.4%), Citrobacter species
(1.4%), and Serratia species (0.9%) in their finger nail. Therefore, food handlers should ensure that their finger nails are
trimmed. Similar types of bacterial isolate were identified among food handlers in other parts of Ethiopia including Jimma
and Gondar . Our finding also goes parallel with different studies carried out in other countries like Nigeria , Iran , Brazil, and
In the present study, approximately one-sixth of cultures of fingernail contents was found to be positive for coagulase-
negative Staphylococci (12.3%) followed by Staphylococcus aureus (5%). Our results are in agreement with previous
studies reported from other parts of the country. and are similar to findings of Zaglool et al. from Saudi Arabia who reported
these bacteria as the most common pathogen isolated from food handlers [9]. Coagulase-negative Staphylococci are the
normal flora of the skin, and this is the reason why high prevalence in this study. In addition, isolation of Staphylococcus
aureus in this study was because it is the true pathogenic bacteria included in the resident microflora of the skin and nose.
Food handlers can easily contaminate food with Staphylococcus aureus (common cause of food poising) if they do not wash
their hands properly after using toilet and after making contact with their nose [22]. Tambekar et al. also reported the
reduction in the number of pathogens after hand washing [23].
Different species of Enterobacteriaceae were isolated in 11.5% of food handlers (data not shown) in the present
study. Klebsiella and Escherichia coli were the predominates, supporting the concept of contamination by fecal bacteria due
to inadequate handwashing by the food handlers, which are a cause of concern for the public [24]. Furthermore, only 15.5%
of the food handlers in our institution had been trained in safe food handling practices.
Escherichia coli is a normal flora usually present in the intestine even though some serotypes (i.e., 0157 : H7) can cause
serious diseases to humans [25]. It is normally absent in hands, and the presence of this bacterium gives a clue of current
fecal contamination with enteric pathogens [26]. Foods that are contaminated with Escherichia coli and Staphylococcus
aureus that do not require further heat treatment could cause food-borne illnesses [27]. Escherichia coli was detected on the
hands of 2.7% of food handlers’ in the current study, which is in accordance with 1.8%–3.9% isolation rates reported in
earlier studies [17, 28, 29]. However, this figure is lower than 22%, 10.9%, and 7.8% carriage reported in Jimma, Iran, and
Turkey, respectively [16, 19, 28]. The difference between our results and the other studies may be attributed to sampling
techniques as well as the different used methods for detection.
Pathogens that can be originated from undercooked or uncooked animal products like Proteus and Klebsiella can
contaminate hands of food handlers from where they could be transferred to foods and the customers [16]. Pseudomonas
aeruginosa is resistant to most antibiotics and disinfectants; hence, when transferred to foods through the nails of food
handlers, food poisoning may occur, and isolating or identifying this pathogen is dangerous [27].
In the present study, no intestinal parasites were detected in the fingernails of food handlers. This finding is in line with the
result obtained from study done earlier in Gondar town, Ethiopia and Makkah, Saudi Arabia [9, 10]. Though, other previous
reports indicated the presence of intestinal parasites in the fingernails contents of study participants [29, 30]. Likewise, all of
the food handlers were not positive for salmonella and Shigella species in respect of their fingernail contents in the present
study, which is also in line with previous studies done in Gondar [10, 17] and Abuja, Nigeria [29].
The Staphylococcus aureus and coagulase-negative Staphylococcus found in the finger nail content were resistant to
multiple antibiotics in this study. Staphylococcus aureus isolated from finger nail contents was resistant to methicillin. If it is
transmitted to patients, it may cause epidemics in patients. The finding of this study is consistent with the previous study
done in Gondar University Cafeteria, Northwest Ethiopia [17].
In the current study, there was significant association between bacterial isolation rate and service years (). This finding is in
line with the result obtained from the previous study done in Debre Markos Ethiopia [13]. However, this was in contrary with
the findings from Addis Ababa and Arba Minch University, South Ethiopia [31, 32] where no statistical significant association
between bacterial isolation and service was seen. This result indicated that food handlers with more work experience have
less risk of bacterial finger nail isolation. This could be explained as food handlers with more work experience have better
personal hygienic practices than inexperienced food handlers [13].
Colonization of bacteria on hands can be facilitated by having untrimmed fingernails because it makes hand washing
difficult and less effective. Wachukwu et al. have showed that food handlers with long nail become colonized and become a
possible risk for transmission of pathogens [27]. In addition, a study conducted by Lau et al. on removal of Escherichia
coli hands with natural or artificial fingernails indicated that untrimmed fingernails tend to carries more microorganisms than
untrimmed nails [33]. Our study also indicated statistically significant association between the isolation of bacteria and finger
nail status ().
In our study, no significant association was found for finger nail bacterial content by sex, age, educational background,
medical checkup, training status, and hand washing habit of food handlers. However previous study conducted in Jimma
indicates significant association between bacterial hand contamination rates with age [16]. Similarly, finding in Sari, northern
Iran, showed that statistical significant association was observed particularly bacterial infestation comparable for educational
level and handwashing practice after using toilet [22]. But this may be due to the small sample size.

5. Conclusion
In general, the present study emphasized the use of different intervention measures that can be used to decrease or
eliminate contamination of foods by food-handlers as well as spread of pathogens to the customers or the public. Therefore,
creating awareness specially on food handling practices and hygienic measures of food handlers is a crucial issue to
prevent the food poisoning pathogens. Specially, individuals with long finger nail harbor food debris, microbial
contaminations, and allergens. Therefore, their use should be under control or supervision by the responsible body in the
institution with much customers.


WHO: World Health Organization.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no competing interests.

Authors’ Contributions

AM was the primary researcher and conceived the idea for this study. AR participated in data collection, conducted data
analysis, and drafted and finalized the manuscript for publication. YA participated in data collection, conducted data
analysis, and drafted and finalized the manuscript for publication. AR and YA read and approved the final manuscript.


The authors acknowledge the financial support by the Debre Markos University College of medical and health sciences.


Nail Properties and Bone Health: A Review

Pouya Saeedi,1 Amin Shavandi,2,3,* and Kim Meredith-Jones4

Author information Article notes Copyright and License information Disclaimer

This article has been cited by other articles in PMC.


1. Introduction
Bone is a metabolically active tissue. Bone remodeling, i.e., bone formation and resorption, continues throughout life. The
imbalance between this coupling process leads to excessive bone loss, little bone formation, or the combination of both.
This phenomenon can cause osteoporosis or porous bone, which increases the risk of bone fracture. Calcium (Ca) and
magnesium (Mg) are important components of human bones, and their deficiencies are associated with the development of
osteoporosis [1,2]. Although medical interventions are indicated and proven to be efficacious with respect to fracture
prevention [3], early diagnosis of osteoporosis is essential.
Bone densitometry is the gold standard for diagnosing osteoporosis and dual X-ray absorptiometry (DXA) is the most
commonly used densitometry technique [4]. DXA technology also incorporates other measurements that have added value
in fracture risk assessment, such as hip structural analysis and trabecular bone score [5]. In addition, DXA can detect
skeletal sites that have low bone mineral density, but it is not able to identify the microstructure of bone, crystal shape and
size, and the interconnectivity of bone porous structure. DXA is also not able to detect microfractures or identify individual
bones at risk of fracture [6,7]. Furthermore, deficiency of the collagen protein in bone structure has been shown to be
associated with bone fracture [8] and DXA is not able to assess the role of proteins in bone health. Therefore, alternative
methods to detect bone architecture are required.
Finger and/or toenails may be good indicators of metabolic changes occurring in the body, as they are in contact with
periosteum of the phalangeal bone. Therefore, the physiological and pathological processes of blood and bone might
influence the nail mineral content [9,10]. Thus, the mineral composition of the nail plates might be a suitable adjunct to bone
densitometry to monitor bone health, i.e., the bone mineral metabolism pattern. Fingernail plates grow at a rate of 3.5 mm
per month and different components including drugs, toxins, and biomarkers are passed into the nail tissue [9]. Nail samples
are also easy to collect, transport, and store [11]. Therefore, nail clippings have been used for the detection of certain toxic
components, exposure to heavy metals [12], nutritional imbalances such as iron deficiency [13], and to examine the
relationship between the concentration of fingernail microelements with coronary heart disease and hypertension [14].
Studies have also evaluated the level of minerals such as zinc (Zn) [15] and selenium (Se) [16,17] in toenails and their
possible association with the risk of myocardial infarction.
In addition to the mineral content of nail, human nail plates also have a three-layered protein structure which includes alpha
keratin, keratin microfibrils in the globular matrix, and keratin associated proteins (Figure 1A) [18]. Keratin is the main
protein in nail, while collagen is the main protein in bone. Both keratin and collagen experience non-enzymatic and post-
translational modifications that can be detected using techniques such as Raman spectroscopy. Post-translational changes
that occur in nail disorders may also be associated with disorders in bone collagen (Table 1) [19,20]. Thus, nail mineral and
protein content might be a useful alternative or complementary method for the screening and detection of bone metabolism
disorders. This review summarises and critically evaluates studies that have examined the relationship between the
physicochemical properties of human nail plates and bone health.
Figure 1
(A) Schematic and cross-section of the nail plate. The nail clipping is normally obtained from the nail plate edge [33]; (B)
Histogram shows disulphide content of nails obtained from 169 women subjects; (C) Disulphide content of nails obtained
from 169 women subjects in relation to age. The figures taken from [31], with permission of Springer.
Atomic absorption spectrophotometer (AAS). Instrumental neutron activation analysis (INAA). Laser-induced breakdown
spectroscopy (LIBS). Dual X-ray Absorption (DEXA). Inductively Coupled Plasma-Atomic Emission Spectrometry (ICP-

2. Association between Nail Mineral Composition and BMD

Calcium (Ca) is the major elemental composition of bone and its deficiency is usually associated with osteoporosis [21].
Magnesium (Mg) also plays an important role in adjusting and controlling bone metabolism. The nail plate is in contact with
the osteogenic layer of phalangeal bone for four to five months and is influenced by pathological and physiological changes
in bone, such as changes in the content of Ca and Mg [10]. Nail also has deposits of Ca and Mg, which can be measured
using techniques such as instrumental neutron activation analysis (INAA) [22]; a technique that is relatively fast and safe.
Using INAA, the possible relationship between Ca and Mg content of toenail and BMD at the index finger from 220 women
aged between 40 and 70 years was measured and shown to be very weak (correlation coefficients ranging from 0.03 to
0.18). The analysis was performed twice in a five-year interval and the authors reported a significant decrease in Ca
density, but not Mg density. Given the weak correlation between the tested nail minerals and BMD, the authors suggested
that the INAA method might not be a sensitive or reliable enough tool for osteoporosis screening [22]. However, in this
study, nail mineral (Ca and Mg) content was compared to radiometrically measured BMD and not to bone mineral content.
Thus, the weak correlation could be explained, because comparisons were made between two un-identical parameters of
nail and bone. The difference could also be due to the site the nail was taken from (toenail) and the site BMD was tested at,
i.e., the index finger. In addition, the radiometric method used to measure BMD in the index finger is considered to be
insensitive, which may partly explain the insignificant relationship [10].
In a study by Bahreini et al. [23], laser breakdown spectroscopy (LBS) was used to measure the elemental composition of
fingernails in 99 subjects (22–89 years), including 27 healthy, 47 osteopenic, and 25 osteoporotic adults. The association
between fingernail minerals and BMD was investigated. Although the results demonstrated a relationship between nail
mineral composition (Ca) and BMD, the validation results were not promising. In the validation sample, only 38.4% of the
cross-validated cases were classified correctly. In addition, there was no correlation between the elemental composition of
fingernails and risk of fracture or osteoporosis [23]. A study in older women (60–69 years) has demonstrated relationships
between Ca and Mg and BMD [10], whereas others have not reported similar findings [11,24].
In a study in elderly women (≥80 years), a significant positive correlation was reported between the Ca to Zn ratio (Ca/Zn) in
fingernails measured using atomic absorption spectroscopy (AAS) and BMD measured with DXA [25]. In addition, the Zn
concentrations in fingernails were significantly but negatively correlated with BMD (r = −0.39). A negative relationship
between Zn concentration and BMD has also been reported in Iranian osteoporotic women aged 36–60 years [24]. The
contradictory findings between studies might reflect the differences in the health status and age of the sample populations,
as well as the analysis techniques in these studies.
Furthermore, higher levels of Zn and a significantly lower ratio of Ca/Zn and Mg/Zn were observed in the nail plates of
osteogenesis imperfecta (OI) patients compared to normal controls. OI is a genetic metabolic disorder that affects the
synthesis of collagen type Ι, where a minor trauma can result in fractures [26]. The high Zn content of fingernails in OI
patients might be related to the disease process, which may increase Zn release from the bone [26]. Therefore, nail
samples could potentially be used for the identification of OI patients. However, the pathomechanism of Zn release from
bone and its accumulation in nails need to be revealed through further studies and for correctly diagnosing OI, the clinical
assessment needs to be coupled with genetic mutation analysis [27].
The lack of consensus between studies may be related to the methodology used for measuring the mineral content and the
population’s age or health status of those tested. In addition to the methods used to analyse nail samples, the variety of
techniques used in the collection of nail samples may explain the discrepant results. Another methodological consideration
is the location the nail sample is taken from. Nails on the thumb and little finger grow slower compared to other fingernails.
Slower growing nails are in contact with the body for a longer time and therefore may have a different mineral content
compared to other fingernails. Future studies should report the finger that has been used for collecting the nail plate
samples. Identifying the finger used to supply the nail sample is not routinely reported in the literature, which might partly
explain the variation in findings.

3. Association between Nail Protein and BMD

Keratin is the main protein constituent of human nail. The cysteine content of keratin determines disulphide (S–S) bonds
and the physiochemical properties of nails. The formation of S–S bonds is essential for the protein’s structural integrity. A
disorder in the metabolism of sulphur can negatively affect the S–S bonds and result in abnormalities in keratin containing
tissues such as nail. The presence of S–S bonds from sulphur containing amino acids (e.g., cysteine and methionine) can
be determined using a technique called Raman spectroscopy. In Raman spectroscopy, the molecules are vibrated after
being subjected to a laser source. The recorded spectra contain information on the vibration of the molecules, which can be
used to differentiate between various Raman active molecules [28,29]. Raman spectroscopy is non-destructive and does
not require specific sample purification or extraction. It is also possible to acquire information about the secondary structure
of proteins such as vibrations of amide I and III [30].
Given the similarities in structure and composition of nail keratin and bone collagen proteins, changes in the nail keratin
properties may be a marker of osteoporosis [29]. As Raman spectroscopy can measure the degree of protein sulphating in
fingernails, the technique may be used to evaluate bone health. In a patented application in the USA, Raman spectroscopy
was used to demonstrate variation in conformations of the S–S bonds at 510 cm−1 and C–S bridges at 625 cm−1 and 641
cm−1, which were then associated with nail and bone health properties. The height of the S–S peak was also demonstrated
to vary between subjects, with an approximately normal distribution (Figure 1B), and it was not associated with subjects’
age (Figure 1C) [31]. The patentee suggested that Raman spectroscopy of nails was a useful diagnostic tool to assess bone
health [32].
The same patent group [34] also evaluated the brittleness and S–S bond content of nails using nano indentation and
Raman spectroscopy. The nail plate samples were collected from women with and without osteoporosis. The nails from
patients with osteoporosis had a lower mean modulus compared with non-osteoporotic women. Additionally, lower S–S
bonds were detected in the group with low BMD. The relationship between a lower content of S–S bonds and higher
brittleness in the tested nails was not discussed. This group also investigated the effect of the reduced content of keratin S–
S bonds on nail mechanical strength. The authors reported that the mechanical rigidity of nail is associated with the nail
protein disulphide content and the fingernails of women with a history of fracture showed a lower disulphide content than
women with no history of fracture (Figure 2A)
Figure 2
(A) Raman spectra from healthy (top) and osteoporotic subjects. The nail samples from osteoporotic patients had a lower
S–S bond content than the healthy group; (B) Relationship between disulphide peak (∼510 cm−1) intensities and the BMD
scores for the (a) hip and (b) lumbar spine groups. The nail Raman spectra arranged based on the osteoporosis presence
in both hip and lumbar regions and compared to normal specimen. As shown, there were no relevant differences between
these groups. The figures are taken from [33,35] with permissions from Springer. (C) The fingernail protein content in the
patients with fracture was significantly lower than other tested groups. Comparison fingernail protein content in
three study groups [42].
The peak of the S–S bonds may also be related to bone health. More than 7% of the total amino acid content of nail is
cysteine, which has a high content of sulphur. Raman spectra analyses of the nail sulphur bondings are in the region of
300–700 cm−1. In particular, the peak at 510 cm−1 indicates the S–S bond. Higher peak intensities at 510 cm−1 have been
reported for healthy nails compared to osteoporotic patients, indicating more cysteine and S–S bonds. In addition to a
greater number of S–S bonds, a C–S spectral shift has been reported in the nails of adults with osteoporosis, which may be
related to the S–S bond cross linking density [29,31]. Other studies [33] have also reported a 25% reduction in nail keratin in
osteoporotic women and peaks of S–S bonds have been demonstrated to be sharper in nails from healthy adults compared
to nails from adults with osteoporosis, indicating the higher content of S–S bonds in healthy nails. All of these studies were
patented studies, and mainly cross-sectional with small sample sizes, which may have precluded detecting changes in the
protein composition of the nail. However, Mussatto et al. [35] have not demonstrated a relationship between the keratin
properties of the nail and bone health (Figure 2B). The discrepant results may be due to the lack of a proper spectra
baseline correction model. Positive relationships have been demonstrated by Beattie et al. [36], who developed a baseline
correction model using a singular value decomposition (SVD)-based method.
Using a baseline spectra correction model, Beattie and associates evaluated the amino acid profile of nail keratin samples
from both fracture and non-fracture samples [36]. The keratin from the fracture group had no evidence of S–S bonds and
had a low content of basic and hydroxyl amino acids compared to the control samples. The fractured samples also had a
higher content of aromatic amino acids. In addition to the observed difference between individual amino acids, the alpha
helices and beta sheets were more ordered and structured in the non-fracture samples compared to the fractured group.
This change in the protein structure could be related to the reduction in the content of S–S bonds. A lower number of S–S
bonds indicates less cross-linking between the chain of the keratin, which leads to a keratin chain with a lower integrity and
Most of the studies on protein properties of the fingernail and its association with bone health have not considered the
relationship between parameters such as the width, area, or intensity of S–S bonds with bone densitometry data. To-date,
one cross sectional study by Cummins et al. [37] has analysed the width of S–S bonds and found no relationship between
Raman spectra data and BMD.
In people with osteoporosis, the activity and number of osteoblasts decrease and the function and activity of osteoclasts
increase. When the bone loses its mineral content, collagenase begins to reabsorb collagen. Under disease conditions such
as osteoporosis, the reduction in the density of bone may indicate a deficiency in the formation of a protein matrix.
Therefore, the reduction in the amount of Ca may reduce the quantity of collagen and S–S bonds in the bone [38]. To
assess this, nail Raman spectroscopy was carried out in a sample of 213 women (61.5 ± 9.7 years) [35]. The nail spectra
and the chemical composition of participants with and without osteoporosis were similar and no differences were seen
between the disulphide peaks. There was also no association between BMD values and the Raman spectra intensity of the
nails at 510 cm−1 or S–S bonds.
Raman spectroscopy may also be a valuable method to identify those at risk of fracture. Two recent studies [36,39] have
collected nail samples from 633 postmenopausal British and Irish women and analysed them using Raman spectroscopy.
Forty-two percent of the tested population experienced a fracture due to fragility. Comparing DXA measured BMD, Raman
spectroscopy, and biomarkers of bone health data, the authors reported Raman spectroscopy as the most accurate method
to differentiate between individuals with and without fracture (p = 0.003). As inter and intra S–S bondings are required for
the synthesising, folding, and stability of collagen, the observed changes in Raman spectroscopy of nail S–S bonds may
indicate changes and disorders in bone. However, as this was a cross-sectional study in patients with existing fracture, the
Raman spectroscopy studies of this kind may not be able to predict risk of future fracture. Furthermore, the absence of
fracture may simply indicate a lack of low impact trauma. In addition, the group with no fracture may not have been
homogeneous in terms of bone health and therefore the results of these studies do not provide strong evidence that Raman
spectroscopy can be used as a diagnostic tool or to predict fracture risk.
Go to:

4. Association between Nail and Bone Proteins

Bone proteins such as osteonectin and signaling proteins of growth factor-B (TGF-B) contain cysteine and S–S bonds.
Osteonectin is vital for bone structure, remodeling, and maintenance [40]. Osteonectin and TGF-B-signalling proteins are
also composed of S–S bond dimers made of two chains of polypeptide that are highly rich in cysteine residues [40].
Cysteine residues and S–S bonds are important for the three-dimensional structure of the bone morphogenic protein 2
(BMP-2), a subset of the TGF-B signalling proteins superfamily [41]. Human bones are made up of either cortical bone or
trabecular bone, and trabecular bone is known to decrease in mineral with age. Given that osteonectin content is 20–40
folds higher in human trabecular bone compared to cortical bone, a nail S–S bond measurement may be a viable method to
measure the health of trabecular bone.
Other bone formation markers including osteocalcin, serum C-terminal telopeptides of type I collagen (serum cross laps),
serum osteoprotegerin, and serum sRANKL have been measured and compared to nail protein in a large sample of
postmenopausal women [42]. Nail protein content was significantly correlated with the concentration of serum cross laps
[42] after controlling for BMD, age, and body mass index (BMI). The nail protein content in patients with fracture was also
significantly lower than both healthy and osteoporotic adults (Figure 2C). Nevertheless, it was not clear how many of the
patients exhibited fracture and some of the individuals with fracture history had zero or negative values for the fingernail
protein content.

Nail Viscoelastic Properties and Bone Health

It has been suggested that the viscoelastic properties of nail may be associated with bone health, as load and time
displacement tests have shown different depths of indentation and higher creep in nails of osteoporotic adults [29]. To
measure the depth of indentation and creep, a maximum load of 2 mN was applied to the nail samples and held for 2000s,
followed by unloading. This loading phase resulted in a viscoelastoplastic response. The nail samples reacted on this load
with a deformation (i.e., creep). While the elastic-plastic loading segment of nails was identical for both healthy and
osteoporotic individuals, the nails from osteoporotic adults showed higher displacement over the tested time. The
conclusion was that nails of adults with osteoporosis showed differences in the time-dependent response to mechanical
load, but not in the hardness nor the elastic modulus. Moran et al. [33] and Pillay et al. [34] evaluated the elastic modulus of
fingernail plates and did not report a significant association between the nail S–S bond content and the nail’s modulus. The
lack of finding may have been because the authors tested the polymeric elastic region of the nail plates, which has both a
viscoelastic and viscoplastic property, but it is the visco-elastoplastic property that is mostly affected by the density of the S–
S bond and not the hardness or elastic modulus properties.
5. Conclusions
Nail samples can be used to measure nail properties such as hardness, elastic modulus, and content of the S–S bonds. Nail
plate sample collection is relatively easy, fast, and inexpensive. Methods to determine bone health using nails show
promise, but studies investigating nail mineral and bone density are hampered by the methodology used to measure these
parameters. Methods such as Raman spectroscopy have been able to detect differences between fracture and fracture-free
groups. However, variations in physicochemical properties of nail and differences in nail growth speed in different
individuals limit measuring bone mineral content through the nail and consequently evaluating bone health through the nail.
Furthermore, studies have not been consistent and the measurement of nail plate protein and minerals cannot yet be
considered as a routine practice for the screening of bone disorders. Given that Raman spectroscopy can provide insight
into bone architecture and not just bone mineral density, it may be a promising investigational tool for the measurement of
overall bone health and fracture risk. However, it is hardly plausible in clinical practice that Raman spectroscopy will be an
alternative to DXA in the clinical assessment of bone health. Future studies should investigate the use of Raman
spectroscopy to directly measure the protein properties of bone through the skin at various body sites.

Author Contributions
A.S. and P.S. designed the study and performed the literature search and wrote the first draft. K.M.-J. critically revised the
manuscript and helped with interpretation of the data. All authors revised the paper critically for intellectual content and
approved the final version. All authors agree to be accountable for the work and ensure that any questions relating to the
accuracy and integrity of the study are investigated and properly resolved.

Conflicts of Interest
The authors declare no conflict of interest.
Wearable Nail Deformation Sensing for Behavioral and Biomechanical Monitoring and Human-Computer Interaction

 Katsuyuki Sakuma,
 Avner Abrami,
 Gaddi Blumrosen,
 Stanislav Lukashov,
 Rajeev Narayanan,
 Joseph W. Ligman,
 Vittorio Caggiano &
 Stephen J. Heisig

Scientific Reports volume 8, Article number: 18031 (2018) | Download Citation

 4368 Accesses
 1 Citations
 146 Altmetric
 Metricsdetails


The dynamics of the human fingertip enable haptic sensing and the ability to manipulate objects in the environment. Here
we describe a wearable strain sensor, associated electronics, and software to detect and interpret the kinematics of
deformation in human fingernails. Differential forces exerted by fingertip pulp, rugged connections to the musculoskeletal
system and physical contact with the free edge of the nail plate itself cause fingernail deformation. We quantify nail warpage
on the order of microns in the longitudinal and lateral axes with a set of strain gauges attached to the nail. The wearable
device transmits raw deformation data to an off-finger device for interpretation. Simple motions, gestures, finger-writing, grip
strength, and activation time, as well as more complex idioms consisting of multiple grips, are identified and quantified. We
demonstrate the use of this technology as a human-computer interface, clinical feature generator, and means to
characterize workplace tasks.


All known, extant primate species possess a flat fingernail if only on the hallux1. This feature is one of the few
characteristics unique to the order Primates. The most recent common primate ancestor is estimated to have lived near the
end of the Jurassic and the start of the Paleocene eras, approximately 65 million years ago2. The structure of the flat nail
has been conserved this long testifying to its usefulness across many tasks and environments. Nails and volar pads are part
of the integumentary system which covers and protects the body3. These structures interface with the environment and are
used to both sense and manipulate. The apical volar pads of the fingers are also part of the somatosensory system,
providing fine touch tactile feedback as well as haptic perception. The volar pads of the hands are specialized to provide a
tough, but flexible gripping surface. These structures are used by humans in gripping thousands of times a day4, for social
touching, dressing, grooming, scratching, food preparation, and eating. Every kind of tool use from teeth brushing, to typing,
to driving a car involves grip. Although in most tasks the volar pads are in direct contact with the object, nails play an
essential function to enhance both gripping and sensing capability by focusing the volar pulp on the object being
manipulated5. Also, the nail plate itself can function as a tool for scratching, picking, cutting, scraping, managing tiny objects
or as a weapon.

The ungula or nail plate is the visible part commonly thought of as the fingernail. Three layers of cornified onychocytes
make up the ungula. Figure 1A is a scanning electron micrograph of a cross-section through the author’s nail in which the
three strata are visible. Onychocytes are anucleated cells which initially arose from the nail matrix or the nail bed6. These
hardened, dead cells are filled with fibrous, translucent onychokeratin proteins7. The dorsal layer provides a hard, smooth
outer covering to the thicker, more flexible intermediate layer. The ventral layer is produced by and bound to the nail bed.
The entire nail plate is roughly 0.6 mm thick in males and slightly thinner in females but becomes thicker with age
regardless of gender6. Non-pathological nail plates have a gentle convex transverse curve along the longitudinal axis of the
fingertip and typically advance about 0.1 mm per day8. Figure 1B contains a confection created by hand registering a
photograph and x-ray, both of one of the author’s fingertip. Work in the context of psoriatic arthritis using high-resolution
MRI imaging and histology has shown that the nail bed and plate are ruggedly and comprehensively bound to
musculoskeletal structures in four ways9. First, collagen fibers arise from the periosteum of the distal phalangeal bone,
especially the spatulate tuft at the distal end and attach to the underside of the nail bed. Second, the collateral ligaments
stabilizing the distal interphalangeal joint extend and connect to the lateral edges of the nail bed. Third, the extensor tendon
attaches to the base of the distal phalangeal bone and continues on to envelop the nail root. Finally, fibers of the extensor
tendon also continue along the dorsal aspect of the distal phalange and join a thickened periosteum and nail bed. These
tendinous and ligamentous attachments transmit tensile stress but not much compressive stress. The deformable fingertip
pulp in direct contact with the lateral edges and the distal end of the nail plate transmits compressive stress but not much
tensile stress. The anatomical structure and interactions between the nail plate, distal phalangeal bone, musculoskeletal
attachments, and fingertip pulp causes nails to deform in complex but repeatable ways during unloaded movement and in
interactions with objects.

Figure 1

Anatomy of fingernail. (A) Scanning electron micrograph of a fingernail cross-section. (B) Index finger photograph and X-ray
of bone structures with connection schematic.
Humans use their fingers extensively in activities of daily living. Because of this, nails are tempting structures to characterize
for insight into these activities. Our primary objective was to evaluate a system to quantify the kinematics of fingernail
deformation and characterize finger and hand movement idioms using a wearable sensor. Identifying grip types, activation
profile, and maximum grip force can provide clinical features relevant to diagnosing degenerative neurological diseases
such as Parkinson’s disease10,11. Hand strength has been correlated with general health-related quality of life12. Other
studies showed that greater hand strength was positively related to cardiac function13, central nervous system health14,
and cognitive performance among people with schizophrenia15. Understanding which tasks require maximum grip force or
finger pressure and how often repetitive actions occur is necessary to identify and ultimately refactor injury producing
workplace activities. Onycholysis, paronychia, and koilonychia are examples of nail disorders that can result from repetitive
microtrauma16. Various forms of tendinopathy and repetitive strain injuries can result from poorly designed or stressful job

In previous work, several groups have investigated the mechanics of fingertip loading and force transmission to better
understand the relationship between workplace tasks and muscle fatigue, pain, onycholysis, and soft tissue injuries (carpal
tunnel syndrome, flexor tenosynovitis, etc.).

The most comprehensive fingertip instrumentation effort was carried out by Reid et al.18 in the context of extravehicular
activity glove design for astronauts. They studied blood perfusion, galvanic skin response, longitudinal and transverse
fingernail strain and pressure relative to force plate and dynamometer data during a battery of static and dynamic tasks.
Fingernail strain during static pressure at various fingertip angles was quantified by Sakai et al.19. This work produced
curves relating fingernail strain in the longitudinal and transverse axes during fingertip compression using three nail-
mounted strain gauges.

To build a biomechanical model of the hand the force response of fingertip pulp to repeated compression was studied in
quasi-static conditions using cameras and force transducers on a plate during a tapping task20. Pulp deformation and
recovery after repeated indention cycles was studied by D’Angelo et al.21, but they did not consider the effects over multiple
days and attributed all results to viscoelastic effects. All the above studies provided insights into fingertip dynamics and nail
deformation through fingertip interaction with surfaces but were limited to bench activities and hence not able to
characterize activity in daily life.

Other authors have explored the possibility of using nail-mounted devices for human-computer interactions. The original
work in this area used nail-mounted cameras to capture differences in the hemodynamic state of the fingertip visible through
the nail during different types of forces. This photoplethysmographic fingernail sensor was used to construct a wearable
mouse that derived horizontal motion, vertical motion, and click behavior from changes in fingertip blood volume via an
array of nail-mounted of LEDs and photodetectors22. Another example, NailO23, prototyped a nail-mounted gestural input
surface that fit entirely on the nail. This device used projected capacitance to detect touch events and communicate
wirelessly with remote computing devices. This device was also limited to simple directional swipe and click gestures similar
to a mouse. Non-existent or occluded input device type controls were presented on a fingernail mounted OLED display in
the NailDisplay24 project. It included an accelerometer to capture basic swipe gestures. Finally, FingerPad25 was a thumb,
and index fingernail mounted device which used magnetic tracking to deconvolve the motion of the thumb on the index
finger to input digits. To the best of our knowledge, no system has integrated both nail strain and accelerometer information
to explore human hand biomechanics and enabled an unconstrained human-computer interaction. Table 1 summarizes the
contribution of our system versus previous projects and devices.

Table 1 Comparison of related work.

We present a wearable fingernail deformation system containing a strain gauge, electronics, and software. This system
characterizes grip, gestures, and more complicated motion idioms. Characteristics of motion idioms such as duration, force,
and fluency describe a human subject’s state. The system also functions as a human-computer interface by translating nail
deformation signals during finger writing to characters. Finger writing can take place on any convenient surface such as a
table, desk, wall, clothing, or subject’s other palm. A finger appropriately configured with a sensor could use gestures to
perform computer operations such as scrolling, paging, shrinking or expanding images. While our ultimate goal is to make a
wearable device that is entirely on the nail and to develop a platform that has the capability of characterizing various kinds
of idioms and gestures, this paper describes a prototype for nail deformation sensing with a wearable device and results of
the demonstration of this concept.


To find optimal locations on the nail for measuring strain forces (see Methods) we initially quantified deformation across the
entire nail during normal force using a three-dimensional Digital Image Correlation (DIC) measurement system (see
Methods). Figure 2A shows three-dimensional surface profile measurements of a fingertip produced by the DIC technique.
When the fingertip pad was pressed against a test surface with normal force, the distal phalangeal bone pulled the center of
the nail downwards, and the fingertip pulp moved around the distal phalangeal bone to push up against the nail at the lateral
folds, deforming the edges upward. The leftmost image in Figure 2A shows displacement, while strain in the fingernail is
shown in the center and right images respectively. The displacement (dL) was more than 130 µm (from −54 µm to +80 µm)
and the strain change was more than 0.3% (from −0.21% to +0.13%). The tests were repeated to understand the
consistency of the DIC measurement and the data remained qualitatively similar. We concluded that a strain sensor
attached to fingernail needs to be flexible enough to tolerate the anticipated deformation (130 µm) and sensitive enough to
detect strain in the range of ±0.20% (=2000 µɛ). From the DIC measurement results, the positions with the most
deformation were across the nail at the top of the lateral nail folds. Figure 2B shows deformation at five newtons of normal
force for a nail with and without a strain gauge glued to it. The profile plots on the right show that the maximum Z-axis
displacement longitudinally was 40 µm without the strain gauge and 10 µm with it. The displacement in the transverse
direction was −7 µm with the strain gauge and −20 µm with it. These results show that when a strain gauge is glued to the
nail it is considerably stiffer and less responsive. Figure 2C shows nail deformation during shear movements. In general, as
finger pulp is pulled away from the direction of motion, it pulls the forward edge of the nail with it.

Figure 2Fingertip surface profile measurements. DIC images of fingernail during normal force test. (A) Left to Right: Z-axis
displacement, strain along X (center), and strain along Y (right) directions. (B) Z-axis displacement and strain profile at five
newtons with and without a strain gauge. (C) Z-axis displacement during shear forces. Hardware Stack Description The
DIC results informed our decision to use a metal foil strain gauge as well as providing the intuition that a low modulus
prosthetic would absorb the amount of nail deformation likely to be seen during normal activities. We used two
configurations during our testing. In the first configuration, the electronics ride on a silicone prosthetic on the nail as shown
in Figure 3A. In the second configuration, seen in Figure 3B the electronics were placed on the skin with a gasket leaving
the strain gauges visible. This visibility allowed us to monitor and document the position and adhesion of the strain gauges
on the nail during testing. Similar to the strain gauges the low modulus silicon prosthetic affected nail deformation. These
effects were consistent, and so it was possible to train models specific to the configuration. Subjects were conscious of
having the device on their nail, similar to wearing a watch. No subject reported discomfort.

Figure 3 Schematic diagram of the wearable system. (A) Electronics with silicone prosthetic on the nail. (B) Electronics
on the skin with for debugging. Schematics of the fingertip model in panel A have been reproduced with permission by
Bucknell Webb. The metal foil strain gauges were rigidly attached to the nail with over the counter cosmetic cyanoacrylate
nail glue (Kiss Professional Nail Glue). We used the three gauge 45-degree rosette (SGD-1/350-RY21, Omega
Engineering) to capture strain readings at different points along the nail rather than to compute the strain tensor. The center
gauge responds to strain in the longitudinal direction only while the outer gauges, oriented at forty-five degrees react to a
combination of both transverse and longitudinal strain.

For each subject, we created a low modulus silicon prosthesis using Skin Tite Silicone Prosthetic Builder, (Smooth-On Inc.,
Macungie, PA, USA). This product is intended for movie special effects and medical prosthetics. The prosthetic adapts a flat
upper surface to mount the rigid boards for the electronics to the surface of the curved nail topology below. The prosthesis
absorbs nail flexion without undue mechanical loading of the nail which could potentially change its behavior or cause nail
problems (e.g., onycholysis). The prosthetic attaches via a gasket to the top of the strain gauge and nail surrounding it. The
gasket is a fabric patch coated with silicon glue (Sil-Poxy Silicone Rubber Adhesive, Smooth-On Inc) on top and medical
adhesive (Pros-Aide Adhesive, ADM Tronics, Inc.) on the bottom. A deformable interconnect is threaded through the gasket
and prosthetic to connect the strain gauges with and board 1 (see Fig. 3A). Two electronics boards sit on top of the
prosthesis. Board 1 contains a 4XX8 BLE 4.2 family programmable system on a chip (PSoC, Cypress Semiconductor, San
Jose, CA). This chip includes an ARM Cortex-M0 CPU and Bluetooth Low Energy (BLE) radio and subsystem. Board 2
contains a 3-axis accelerometer (ADXL335 IMU Analog Devices, Norwood MA) and a coin battery holder. At the top of the
stack a 1.55 volt, 80 mA, silver oxide, high drain battery (Energizer 393/309) fits into the battery holder. On the PSoC, strain
and acceleration values are captured at 50 Hz, analog to digital conversion performed, and the resulting values joined with
an oscillator value. The oscillator runs at 32.768 kHz, and values reset every time the PSoC initializes. This value provides a
way to determine if observations are lost, or the PSoC has been interrupted or restarted and is also used to space the
incoming data properly in time sequence since it will be received in clumps and delayed non-deterministically by the
Bluetooth protocol. The X, Y, and Z accelerometer values, and right, center, and left strain gauge values are all digital
versions of continuous values after ADC processing. The strain values correspond to the strain gauge resistor divider
voltage amplified via a two-stage inverting amplifier op-amp circuit. The distance from fingernail to wrist is on the order of six
inches which makes it too far for near field communication (NFC), but well within the range of Bluetooth low energy (BLE).
Since it is known that the distance from the device to watch will always be in this range, power settings can be reduced. For
testing and debugging, we used a USB dongle plugged into a laptop running MATLAB scripts to plot and record the data.
For naturalistic task testing, we used a Series 3 Apple watch (Apple Inc, Cupertino CA, USA) worn on the wrist to
communicate with the PSoC to receive the data tracks. The raw data was sent via a paired iPhone on to cloud-based
machines for retention, analysis and model training.

Nail strain measurement under basic forces

We gathered information about how the nail deforms under strain responses during some simple actions before
investigating less restricted tasks. In all cases, the device was positioned on the index finger (either left or right, but
independent of the dominant side). Figure 4 shows example responses for a representative subject to presses against a flat
surface with a perpendicular force with a flat finger, finger oriented forty-five-degree angle, sliding motions, and lateral
movements respectively. Next, Figure 5 shows examples of responses during power and precision grips.

Figure 4 Examples of strain response data. (A) Strain response data for a series of flat normal force presses against the
surface of a dynamometer (left) and strain vs. force curves for these same actions (right). (B) Strain response for a series of
normal force, forty-five-degree angle finger presses on a glass surface with the finger pulp first distally and then proximally
displaced (see Supplementary Movie 1). (C) Responses when the finger is pushed away and the onychodermal band pulls
down on the distal end of the nail imparting bending (tensile) stress. (D) Strain responses for a series of sliding motions
toward from the subject under a normal force on a glass surface producing compressive stress. (E) Strain responses to a
series of left and right lateral movements under a normal force on the glass surface. (F) A set of three finger extensions
followed by three flexion movements.

Figure 5
Examples of strain responses during precision and power grips. (A) Strain and force responses during a set of precision
pinch grips. (B) Strain versus force responses during precision pinch. (C) Responses for a power grip session. (D) Strain
versus force responses during power grip
On the left in Figure 4A strain response data for a series of normal force presses against the surface of a dynamometer
(NEULOG NUL-237, resolution: 0.1 N) and USB module (NEULOG USB-200) as shown in the inset photograph, is plotted
with the corresponding force values. On the right in Figure 4A are the strain vs. force curves for these same actions. The
signal saturates for the right and center gauges meaning that at a certain point more force was applied but the nail did not
continue to deform. Fitting a polynomial to the center strain response gave an intrasession r-squared 0.97 for this subject
and an intersession r-squared of 0.84.
Figure 4B shows the strain response for a series of forty-five-degree angle finger presses on a glass surface with the finger
pulp first distally and then proximally displaced. There has been some discrepancy in previous work19 concerning the
polarity of strain in response to normal force. We demonstrate here that the disposition of the finger pulp from distal to
proximal relative to the distal phalangeal bone can switch the polarity of strain from compressive to tensile during normal
force. Supplementary Movie 1 is a recording of this and the next four tests.

Figure 4C shows strain responses for a series of sliding motions away from the subject under a normal force where the
onychodermal band pulls down on the distal end of the nail and imparts bending (tensile) stress. In these movements, we
see that since the center strain gauge is aligned to respond to changes along the longitudinal axis and the gauges at the
lateral edges are at a forty-five-degree angle they only produce half the response. Figure 4D shows the strain responses for
a series of sliding motions toward the subject under a normal force on a glass surface. These produce longitudinal shear
forces which displace the finger pulp distally, and the free edge of the nail is pushed upwards, compressing (or
straightening) the nail in the longitudinal axis.

Figure 4E shows strain responses to a series of left and right lateral movements under a normal force on the glass surface.
The first five movements are left to right, the second four are right to left, and the final four are left to right. In each case, the
leading, lateral side of the fingernail and the center are under compressive stress, and the trailing edge is under tensile
stress during these movements.

Figure 4F shows a set of three finger extension followed by three flexion movements. In the case of extensions, the
extensor digitorum muscle contracts to put tension on the extensor digitorum tendon which terminates on the base of the
distal phalangeal bone and attaches to the proximal end of the nail bed. Tension from this tendon on the proximal end of the
nail bed anchored by central and distal connections to the distal phalangeal bone results in tensile strain on the nail plate as
measured in this test. These examples illustrate structures other than finger pulp acting on the nail. In this case, there was
no contact with any external object, and only musculoskeletal connections were involved.

These tests demonstrate that the sensor system can capture the orientation, direction, and force of interactions at the
fingertip. Due to the anatomical links between the fingernail and the musculoskeletal system, it is also possible to capture
differential forces produced by movements that have no object interaction. We showed this in Figure 4F with a set of three
finger extensions followed by three flexion movements.

Next, we demonstrate the nail sensor characterizing grip related actions. Figure 5A shows the strain and force responses
during a set of precision pinch grips. In Figure 5B the strain versus force responses are plotted. It should be noted that the
center strain gauge did not saturate even at maximum grip force.

Figure 5C shows the responses for a power grip session. In a precision grip all the grip force is transmitted through the
index finger pad, while in a power grip, the entire length of all the fingers, and the palm contribute to the grip force. In
Fig. 5D none of the strain gauges saturated and the force generated by the power grip was almost an order of magnitude
higher than the precision pinch. With a strain gauge on a single finger, we were unable to predict grip type. Consequently,
we could not predict the overall grip force without knowing the grip type.

Nail-deformation sensor for dexterous movements and human-computer interaction

To explore using nail strain measurements as a human-computer interface and demonstrate identifying subtle finger
movement idioms, we attempted to predict digits ‘written’ with a fingertip on the screen of a tablet (see Methods, see
Supplementary Movie 2). Subjects recorded digits during four sessions on four different days. We used a new strain gauge
for each test to eliminate the possibility of damage during removal effecting subsequent tests. Two sessions were used for
training and the remaining two for test and validation. Each digit was represented as a time series of strain values (see
Supplementary Fig. 1). As a demonstration of how unique this data was for each digit, we created an unsupervised t-
distributed stochastic neighbor embedding (t-SNE) plot of the data for all four sessions from a single subject. The purpose of
t-SNE analysis is to show that the continuous high-dimensional time series can be reduced to a lower dimensional space in
which the digits are clearly separated. Indeed, Figure 6A shows the results when the t-SNE technique has been applied to
reduce the data to two dimensions. In this representation, each data point is colored according to its actual digit label. There
are well-separated clusters for each digit.

Figure 6
Detection of finger movement idioms. (A) T-SNE visualization showing separation of strain signals color coded according to
their true values. Confusion matrix of correctly and incorrectly predicted finger digits for the test set. The absolute number of
trials is reported
To predict digits from strain voltage readings we used a 1-nearest neighbor algorithm with the Procrustes distance as the
distance metric. The generalized Procrustes technique overcame the changes we observed in nail deformation response
over time and find the optimal superposition between shapes. Using the first two sessions for training and the second two
for testing we achieved an accuracy of 98.7% for this subject. Figure 6B contains the full confusion matrix. We observed
that over several sessions of finger writing, subject’s fingers became less responsive. We suspect that this was from
loosened connections within the fingertip, possibly the anterior tendon but we are unable to prove this.

To understand other behaviors that might be identifiable from nail deformation but not from wrist-worn devices we captured
data for turning a key, turning a doorknob, screwing and unscrewing a nut, using a screwdriver, and rest. We collected three
sessions of labeled training data consisting of one to three minutes of each task on three different days. When the first two
sessions were used as training data and the third for testing, we achieved an accuracy of 91%. Figure 7A shows a plot of
the task probability prediction for each of the activities during the validation session while Figure 7B shows the complete
confusion matrix for each window prediction. During an additional trial containing all tasks in a random sequence, strain and
video data were collected. Supplementary Movie 3 contrasts the predictions of the continuous activity detection models (for
each 1.5 s window) with the actual performance of those tasks.

Figure 7
Detection of hand activities. (A) Time series of probability activations corresponding to five actions: using a screwdriver,
turning a doorknob, turning a key, screwing and unscrewing a nut, and rest. (B) Confusion matrix of correctly and incorrectly
predicted hand activities for the test set. The absolute number of trials is reported.
Finally, we explored the ability to capture data during activities which produced large nail deformations and required
strenuous grips without interfering with these actions. As a proof of concept, we monitored the signal during a series of
baseball grips and simulated pitches (see Supplementary Movie 4). We reliably received qualitatively specific signals for
each grip. These biomechanical insights into finger forces during specialized, highly energetic idioms were recorded without
inhibiting the subject’s ability to throw the ball or damaging the sensor.

Discussion In this paper, we demonstrated the first of its kind prototype wearable system to capture and characterize
fingernail deformation forces. The system is small enough to be worn on a finger, does not interfere with tactile sensing or
haptic perception, and is sturdy enough to wear while throwing a baseball. Fingernail deformation forces were used to
identify digits written with the fingertip as an example of characterizing subtle finger movements. Finger writing is a simple
demonstration of the system as a human-computer interface. We also quantified the relationship between deformation and
various grip types and grip force.

Since we used rigid printed circuit boards for the prototype, future hardware improvements will focus on shrinking the
electronics using wafer-level chip scale packaging, switching to a flexible substrate and reducing power by changing to
micrometer scale. Replacing the large metal foil strain gauge package with tiny silicon strain gauges will allow the nail to
move more freely. Packaging and installation present many challenges, as the system should ultimately be waterproof and
easy to install and remove. Capturing phenomenology during disease states with known correlations to changes in grip
force is a priority. Finally, identifying and quantifying injury producing motions for immediate feedback to the wearer is a


Data collection All experiments were performed in accordance with institutional guidelines and regulations. The
experimental protocol was approved by the IBM Research Institutional Review Board, at the T.J. Watson Research Center,
Yorktown Heights N.Y., USA. All participants gave their informed consent to take part in the experiments. Four subjects
were involved in finger pressures under basic forces. Three subjects participated in the finger writing tasks and four subjects
in the hand actions experiment. One experienced subject executed various baseball pitch grips.

Measurements of nail forces

Stress is defined as the force acting on a cross-sectional area of a deformable object. Strain is the response to stresses
applied to a deformable object, usually defined as the amount of deformation in the direction of the stress divided by the
initial length of the object, resulting in a unitless number. Strain (ɛ) is defined as ε=dL/Lε=dL/L where L is the length of the
object. In a simple, homogeneous material object, stress and strain have a linear relationship in their elastic range obeying
Hooke’s law. The fingertip is not a homogeneous object but is composed of many elements (bone, tendon, ligament, nail,
pulp) with different elastic properties. The sensitivity of a strain gauge, known as the gauge factor (GF) is defined as the
ratio of change in electrical resistance to the strain GF=(dR/R)/(dL/L)=(dR/R)/εGF=(dR/R)/(dL/L)=(dR/R)/ε where R is the
electrical resistance of strain gauge. The gauge factor of a typical metal foil strain gauge ranges between 2 to 6 and strain
up to at least 10% can be measured. For example, a strain gauge with a gauge factor GF = 2.1 will exhibit a change in
electrical resistance of 2.1 × (2000 × 10−6) = 0.41%, to support strain of 2000 µɛ. By applying strain gauges to the nail plate,
we measure changes in resistance caused by nail plate strain. An additional consideration when using metal foil strain
gauges is the amount of current required. In general silicon strain gauges draw less current but are more delicate and
require additional packaging invention.
Quantification of nail deformation
Nail strain forces were quantified as deformation across the entire nail during normal force using a three-dimensional Digital
Image Correlation (DIC) measurement system (Aramis − 3D Motion and Deformation Sensor, GOM). This system uses two
cameras in a stereoscopic arrangement with a known distance to the object to calculate the position of each surface point.
Finding the maximum correlation between the pixel values of small regions on the surface of the object in two different
images results in a set of transformation parameters. These transformation parameters are then used to compute
deformation and displacement for each region. Strain can be computed from the transformation parameters and changes in

Finger idioms Subjects recorded digits during four sessions on four different days. A new strain gauge was used for each
test to eliminate the possibility of damage during the previous removal having an effect on subsequent tests. During each
session, up to forty instances of each digit were recorded. An app was used (FieldNotes) to capture which digit was written
and when to provide labels for the training data. This labeled data was separated into time series snippets containing the
analog-to-digital (A2D) voltage values returned for the three strain gauges while each digit was written. A bandpass filter
(from 0.25 Hz to 3 Hz) was applied to the data and the absolute value computed and summed back into the original data.
This signal was used to separate each digit in the time series (see Supplementary Fig. 1). The strain data for each digit was
padded to form a three by three hundred element matrix, and the first twenty principle components were computed and
passed to the t-SNE algorithm. Each data point was colored according to its true digit label.

Object manipulation In order to capture key and doorknob turning, nut screwing and unscrewing, and using a
screwdriver, one session of labeled training data consisting of six minutes of each task. A final session containing all four
tasks (in addition to rest) in random sequences was collected, and video was collected. In this experiment, both strain and
accelerometer data were used. All data tracks were resampled to 50 Hz and bandpass filtered (from 0.25 Hz to 15 Hz) to
remove the offset and high-frequency noise. Each sample was a matrix of six tracks (three strain and three accelerometer)
and seventy-five elements corresponding to a 2-second window that was then moved by 0.25 seconds to form the next
sample and was labeled by task. The classification model was a neural network with a hybrid CNN/LSTM architecture26. An
architecture diagram is included in the Supplementary Figure 2 and consists of one Convolutional layer, a max pool layer,
an LSTM, two dense layers and a soft-max layer. All layers are fully connected. The inputs (size 6 × 96) are first processed
by a 1D convolutional layer with 100 convolutional filters (kernel size = 5, stride length = 1, linear activation function)
followed by a max-pooling layer (size 2) to reduce feature representation from 96 to 48. Convolution is done across time
and its resulting feature maps are fed to an LSTM (150 units, linear activation function), then a sequence of two dense
layers (100 units and 5 units, linear activation function). Finally, the soft-max output layer computes probability activations
over our 5 action states. The model was trained to minimize the cross-entropy loss function using the Adam optimization
algorithm. Regularization was done by means of dropout (30% rate before the Dense layers) to reduce overfitting. All the
code was developed in python using the Keras library.

Common nail changes and disorders in older people

Diagnosis and management

Lina Abdullah, RN
Ossama Abbas, MD
Author information Copyright and License information Disclaimer

This article has been cited by other articles in PMC.

Older people are at an increased risk of nail alterations, including normal age-related changes and disorders that more
commonly affect this specific population. Secondary factors are important contributors to pathologic nail changes, including
impaired circulation at the distal extremities, faulty biomechanics, infections, neoplasms, and skin or systemic diseases with
nail manifestations.1,2 These factors can affect primarily the nail plate or involve other components of the nail unit such as
the matrix, nail bed, hyponychium, or nail folds (Figure 1), with secondary abnormalities in the nail plate. These nail changes
can either cause serious symptoms, impairing the daily activities of this older population whose activities might already be
restricted, or be asymptomatic but associated with substantial cosmetic problems, leading to negative psychological effects.
A primary care physician who is knowledgable about and familiar with these age-related nail alterations and disorders will
be able to recognize and manage common pathologic changes, as well as refer patients for more specialized care, if
Figure 1.Anatomy of a nail unit
Quality of evidence
Using a literature search and cross-referencing, we identified articles published before September 2009 that were relevant
to the topic. In September 2009, we performed a MEDLINE search using MeSH terms nails diseases and aged with key
words relevant to each specific age-related nail change and disorder. We identified 2496 articles, 32 of which were selected.
We mainly chose randomized controlled trials, meta-analyses, and review articles, when available, on each specific nail
entity, especially those concerned with elderly patients. When such strong evidence was not available, which was the case
for some conditions such as scabies involving the nail unit, case reports or series were chosen. We selected only those
articles written in English. Using the 3-point grading classification system of evidence-based medicine and given the paucity
of evidence on some of the conditions, articles with different levels of evidence (I, II, or III) were included in our review.
:Age-related nail changes
With advancing age, normal characteristic changes in the growth rate and morphology of the nail plate occur. 1,2 The
underlying mechanisms for these changes are still not completely understood but might be related to dysfunctional blood
circulation at the distal extremities or to the effects of ultraviolet radiation.
Nail plate growth rates of fingernails and toenails normally average 3.0 and 1.0 mm/mo, respectively. With advancing age,
starting at the age of 25 years, this rate tends to decrease by approximately 0.5% per year.1,2
Age-related changes in the morphology of the nail plate include alterations in its thickness, contour, surface, and
colour.1,2 Men generally have thicker nail plates than women do; the normal average thickness of fingernails and toenails is
0.5 and 1.38 mm in women and 0.6 and 1.65 mm in men, respectively. With advancing age, various changes in nail plate
thickness might occur, becoming thicker, thinner, or remaining the same.1,2 A decrease in the longitudinal curvature and an
increase in the transverse convexity characterize senile changes in the contour of the nail plate. 1,2 As for texture, there is
usually a tendency of the normally smooth nail plate texture to become progressively more friable with increasing age,
resulting in fissuring, splitting, and longitudinal superficial or deep striations.1,2 Among nail plate colour changes in elderly
people, the most commonly observed alteration is a yellow to gray discoloration with dull, pale, or opaque appearance. A
peculiar discoloration observed in around one-fifth of people older than 70 years of age is “Neapolitan nail,” which is
characterized by an absent lunula in addition to 3 horizontal bands of white (proximal), pink (middle), and opaque (distal)
discolorations.1 One study found that osteoporosis and thin skin were significantly associated with this peculiar nail
alteration (P < .01) and suggested collagen abnormality as the cause of these changes in nail bed, bone, and skin. 3 Terry
nail, an apparent leukonychia characterized by a proximal white band and distal transverse pink band, is usually seen in
liver cirrhosis and chronic congestive heart failure, but recently it has been observed as a nonpathologic change of the
normal aging process.

Age-related nail dystrophie

Many nail disorders (Table 1)1,2,5–18 affect the population in general; however, they might appear with increasing frequency
with advancing age and include brittle nails, onychauxis, onychocryptosis, infections (especially onychomycosis),
onychoclavus, subungual hematoma, splinter hemorrhages, and malignancies of the nail apparatus. 1,2,5–27
Brittle nails (fragilitas unguium)
Brittle nail disorder is considered a polymorphic abnormality characterized by increased fragility of the nail plate (Figure 2). It
affects around 20% of the population, with increased incidence in women and in older people. 5 It manifests clinically with
varying severity of onychoschizia or onychorrhexis.5 Onychoschizia is usually caused by impairment of intercellular
adhesion between the corneocytes that make up the nail plate. This results in transverse splitting due to breakage of the
lateral edges of the nail plate and in lamellar splitting of the free edge and distal portion of the nail plate. Exogenous factors
(eg, repetitive cycles of wetting and drying, trauma, and fungal proteolytic products) and chemicals or cosmetics (eg, cuticle
removers, nail enamel solvents, and nail hardeners) are among the underlying causes. Onychorrhexis, on the other hand,
frequently manifests as nail plate splitting or ridging, longitudinal thickening, or multiple splits leading to triangular fragments
at the free edge. It is usually the result of nail matrix involvement leading to abnormalities in epithelial growth and
keratinization. Among the various factors causing onychorrhexis are abnormalities of vascularization and oxygenation (such
as anemia or arteriosclerosis), as well as systemic (metabolic, endocrine, etc) and dermatologic diseases (disorders of
cornification and inflammatory diseases)
Figure 2.
Mild brittle nails with mild longitudinal ridging and mild distal lamellar splitting
Management of brittle nail disorder might not be as easy or simple as would be expected.5 Determination of the
predominance of either onychoschizia or onychorrhexis should be the initial step, followed by an effort to identify and, if
possible, correct any of the underlying factors. General and specific therapeutic measures might then be followed. This
includes nail hydration with daily 15-minute soaks using emollients rich in phospholipids (level III evidence).5 Application of
nail hardeners containing formaldehyde can be used to strengthen the nail plate (level III evidence) 5; nevertheless, caution
should be entertained when using these products, as they might cause brittleness, subungual hyperkeratosis, or
onycholysis (ie, separation of the nail plate from the underlying nail bed). Mechanical nail plate protection and fracture filling
can be accomplished using enamel; however, considerable dehydration might occur when it is removed afterward. Several
studies have shown a daily oral intake of 2.5 mg of biotin for 1.5 to 15 months to be of some benefit; however, these studies
were not large or double-blind, placebo-controlled trials (level II evidence).5–7 Finally, one study showed that a daily dose of
10 mg of silicon, in the form of choline-stabilized orthosilicic acid, might be beneficial for treating brittle nail syndrome (level
III evidence).8
Localized hypertrophy of the nail plate (Figure 3), also known as onychauxis, commonly manifests clinically as loss of nail
plate translucency, discoloration, and often subungual hyperkeratosis. It might be associated with pain, and with time can
become complicated by distal onycholysis, subungual hemorrhage, subungual ulceration, or increased risk of
onychomycosis.1,2 Advancing age or faulty biomechanics, which are usually more common in the elderly population (eg,
overlapping and underlapping toes; foot-to-shoe incompatibility; or digiti flexi characterized by contracted toes secondary to
buckling of toes induced by shortening of the controlling muscles), might be contributing factors.
Figure 3.
Onychauxis: Localized hypertrophy of the nail plate with discoloration and loss of nail plate translucency.
Periodic debridement of the thickened nail plate, either partially or totally, is the preferred initial therapy (level III
evidence).1,2,9 Other treatment options that might be of benefit include electric drills, nail avulsion, or 40% or higher urea
pastes (level III evidence).1,2,9 Chemical or surgical matricectomy might be used as a last resort in complicated cases or
those with recurrences in order to achieve permanent ablation of the involved nail plate.
:Onychoclavus (subungual corn)
Onychoclavus is another hyperkeratotic process commonly observed in the elderly. It is typically located under the distal nail
plate margins (most commonly the great toenail), presents as a tender dark area, and can be easily confused with benign or
malignant subungual melanocytic lesions.1,2 Underlying causes include chronic minor trauma and persistent localized
pressure secondary to bony abnormalities such as foot-to-shoe incompatibility, digiti flexi, rotated fifth toes, or hallux valgus
(ie, the great toe rotates toward the second toe).1,2 Treatment includes surgical removal of the hyperkeratotic tissue, as well
as correcting any underlying bony abnormality (level III evidence).1,2
Different pathogens (fungal, bacterial, or even parasitic) can infect the nail plate either primarily or through involvement of
structures such as the nail folds, with secondary extension to affect the nail plate.1,2,10
Onychomycosis is a fungal (dermatophytes, yeasts, or nondermatophyte molds) infection of the toenails or fingernails. It is
by far the most common nail infection, representing around half of all nail diseases and affecting 10% to 20% of adults,
particularly the elderly.10,11 Increased risk of onychomycosis is associated with multiple factors, including male sex, old age,
smoking, underlying medical diseases (eg, peripheral arterial disease, diabetes, and immunodeficiency), and predisposing
genetic factors.10–13 Dermatophytes, mostly Trichophyton rubrum and Trichophyton mentagrophytes, cause more than 90%
of onychomycosis cases, while yeasts such as Candida and nondermatophyte molds such as Scopulariopsis brevicaulis are
responsible for the remaining cases.10
Clinically different subtypes of onychomycosis are recognized, among which distal and lateral subungual onychomycosis is
the most common (Figure 4).10,11 This subtype of onychomycosis, usually caused by T rubrum, presents with onycholysis,
subungual hyperkeratosis, nail thickening, and discoloration, and is caused by fungal invasion that starts initially at the
hyponychium, with progressive proximal spread along the nail bed. Another subtype is superficial onychomycosis (Figure 5),
which presents as black (caused by dematiaceous fungi) or white (caused by T mentagrophytes) patchy discoloration of the
nail plate secondary to fungal invasion of the dorsal surface of the nail plate. 10 Proximal subungual onychomycosis, usually
caused by T rubrum, clinically manifests as a white area under the lunula that progresses distally; it is an important subtype
to recognize, as it commonly affects immunocompromised individuals and can be a clue to HIV infection. 10–12 It results from
fungal invasion of the proximal nail fold with secondary extension to the nail plate. Total dystrophic onychomycosis is also a
subtype of onychomycosis and might be observed in immunodeficient patients, including patients with HIV and those with
chronic mucocutaneous candidiasis.10,13 It is an advanced form characterized by progressive destruction of the nail plate,
leading to the exposure of an abnormally thickened nail bed, and it might be fairly acute or progressive, simply representing
an end stage of other forms of onychomycosis
Figure 4.
Distal and lateral subungual onychomycosis: Onycholysis, subungual hyperkeratosis, as well as nail thickening and
Figure 5.
superficial onychomycosis: White patchy discoloration of the dorsum of the nail plate.
Effective onychomycosis treatment requires reaching an accurate diagnosis with identification of the underlying
pathogen.10,14 Several diagnostic methods such as histopathology with periodic acid-Schiff, KOH-based microscopy, and
fungal cultures can be used alone or in combination; the former being the most sensitive, with sensitivity reaching 98.8%
(level I evidence).10,14 Therapies include antifungal agents (topical or oral), mechanical or chemical treatments, or a
combination of these; the choice of which should be individualized depending on multiple factors such as number of nails
involved, severity of onychomycosis, causative agents, drug side effects, potential for drug interactions (especially in older
persons who are usually already on multiple medications), and cost.15 Terbinafine, whether in continuous or intermittent
regimens, currently appears to be most effective among oral agents for treating onychomycosis caused by dermatophytes
(level I evidence), especially in older patients, owing to its fungicidal effect, safety, and low potential for drug
interaction.15,16 The azoles (fluconazole, ketoconazole, itraconazole) can also be used but, given their fungistatic effect, are
generally less effective than terbinafine.15
Paronychia, seen infrequently in older persons, is an acute or chronic nail fold infection that might result in secondary nail
plate changes.1,2,17 Acute paronychia, typically induced by trauma, most commonly presents as tender erythematous nail
fold swelling of one finger and is usually caused by Staphylococcus aureus. Warm saline soaks, abscess drainage, or
topical or systemic antibiotics are usually used in its management (level II evidence). 1,2,17 Chronic paronychia, on the other
hand, manifests as erythematous and swollen nail folds with cuticle loss and secondary changes in the nail plate in the form
of multiple transverse ridges (Figure 6). Candida species or Gram-negative bacteria are the usual pathogens. Management
includes keeping the nail fold dry in addition to the use of topical antifungal or antiseptic agents (level II evidence). 1,2,17
Figure 6.
Chronic paronychia: Red and swollen nail folds, cuticle loss, and secondary nail plate changes.
Sarcoptes scabiei infestation in specific populations such as infants, immunosuppressed patients, and older people might
have peculiar, uncommon presentations including nail involvement. 1,2,19 Through inhabiting and persisting in subungual
hyperkeratotic debris, the mite can cause prolonged infestations, which might lead to epidemics in nursing homes among
elderly patients and those caring for them.1,2,19 Management includes the use of an antiscabetic treatment coupled with nail
cutting and brushing nail tips with a scabicide (level III evidence).1,2,19
:Onychocryptosis (ingrown toenail)
Onychocryptosis occurs when the nail plate penetrates into the adjacent lateral nail fold secondary to nail plate
overcurvature, subcutaneous in-growing toenail, or lateral nail fold hypertrophy. It manifests clinically with inflammation of
the lateral nail fold, which might be associated with granulation tissue and secondary infection. Although more common in
young adults, onychocryptosis might infrequently be encountered in older persons, resulting in substantial pain, walking
difficulties, and disability.1,2,18 Underlying causative factors include inappropriate nail cutting, long toes, prominent nail folds,
ill-fitting or high-heeled shoes, hyperhidrosis, and bony abnormalities.
Management should address the acute signs and symptoms, as well as correct any underlying predisposing
factors.1,2 Conservative management includes soaking the foot in warm water and placing cotton wisps under the ingrown
edge of the nail plate. Complete cure can best be accomplished by partial nail avulsion and lateral matricectomy, using
phenolization or direct surgical excision of the nail matrix (level I evidence).1,2,18,21
Vandenbos and Bowers, who believe that ingrown toenails might be caused by overgrowth of surrounding skin rather than
nail abnormality, performed a procedure (known as Vandenbos procedure) that was based on excising and removing an
adequate amount of soft tissue from around the nail plate.22,23 However, this procedure, reported as not having any
associated recurrence or osteomyelitis, is not recommended for elderly patients who commonly have associated dystrophic,
thick, discoloured, or curling nails or fungal infections.22,23 Postoperative complications include nail bed infection,
recurrence, or poor cosmetic outcome. Noël18 described surgical decompression of the ingrown toenail (by removing a large
volume of soft tissue around the nail plate and relieving the inflammation) without matricectomy as very effective (level III
evidence).21 Using this method, complete preservation of nail anatomy and function can be achieved with excellent
therapeutic and cosmetic results.
:Subungual hematomas
Subungual hematomas (Figure 7) are commonly observed in elderly people. Subungual hematomas initially present as
painful, red, subungual discolorations that move forward and tend to become bluish and less tender with time. Occasionally,
distal onycholysis with subsequent spontaneous avulsion of the nail plate can occur as a late consequence. It is in fact this
forward and distal movement of the discoloration under the nail plate that can be a very helpful clinical clue, distinguishing
this lesion from melanocytic lesions such as nevi and melanomas. In difficult situations in which an evident history of trauma
is not present, a urinalysis reagent strip can be a non-invasive and very efficient method of diagnosing blood under a nail24;
however, it should be kept in mind that blood presence under the nail plate does not rule out a concomitant neoplasm
completely, as subungual tumours might spontaneously bleed or might be preceded by or first recognized only after a minor
trauma.25 The condition is most commonly caused by trauma, which results in nail bed laceration followed by accumulation
of blood in the nail plate.1,2 Other less common causes include diabetes mellitus, amyloidosis, or anticoagulant therapy.
Management mainly centres on reassurance and observation of the nail; however, in acute, tender cases, relieving pressure
might be accomplished by drilling a hole through the nail plate. After ruling out melanoma, chronic cases are best observed
for spontaneous healing to occur.
Figure 7.
Subungual hematomas: Red to bluish subungual discoloration.
:Splinter hemorrhages
Splinter hemorrhages usually present as linear discolorations under the nail plate, progressing from an early red to a dark-
brown or black colour in a period of a few days.1,2 Splinter hemorrhage location under the nail plate might be a clue to the
underlying cause, as those located in the middle or distal third of the nail plate are typically trauma-induced, while proximal
location is usually associated with systemic diseases such as cholesterol emboli, connective tissue disorders, or infective
endocarditis. The latter proximal type is generally more common among young adults and requires treatment of the
underlying systemic disorder, whereas the former trauma-associated distal type is observed more frequently in the elderly
population and commonly resolves spontaneously.
Malignancies of the nail apparatus
The incidence of common nail apparatus malignancies such as Bowen disease and melanoma tends to increase with
advancing age and is usually highest in the elderly population.
Bowen disease of the nail unit usually originates from the nail fold epithelium, and multiple factors have been implicated in
its pathogenesis, including trauma, arsenic, x-ray exposure, chronic paronychia, and human papillomavirus infection
(especially human papillomavirus 16, 34, and 35).26 It commonly affects the fingers, particularly the thumb. While its usual
presentation is as a periungual or subungual ulcerated hyperkeratotic lesion that might be associated with onycholysis,
other less common manifestations include longitudinal melanonychia (Figure 8) or erythronychia (Figure 9).26 Local invasion
with underlying bone involvement occurs in less than 20% of patients, and the rate of distant metastasis is usually much
lower. The treatment of choice for this condition is Mohs micrographic surgery.26
Figure 8.
Longitudinal melanonychia: A longitudinal dark band can be a manifestation of melanocytic lesions, drugs, or Bowen
disease, among other causes.

Figure 9. Longitudinal erythronychia: A longitudinal red band can rarely be a manifestation of Bowen disease.
Nail apparatus melanoma (NAM) usually affects Japanese and African Americans and classically presents as a solitary
longitudinal melanonychia of the big toe, thumb, or index finger.27 Hutchinson sign, which is characterized by pigment
extension from the nail bed and matrix to the surrounding tissues and which accounts for the radial growth phase of this
melanoma, might also be present. Delay in the diagnosis of NAM might account for its relatively worse prognosis compared
with its cutaneous counterpart. Initial management starts with a high index of suspicion, especially when confronted with an
elderly patient presenting with an isolated longitudinal melanonychia.27 After histologic confirmation of NAM, treatment is
then customized based on the melanoma stage.
:Other nail conditions
Several other conditions should be considered when evaluating an elderly patient with nail changes, including those
changes associated with cutaneous inflammatory disorders (such as psoriasis), 28 nail cosmetics,29,30 systemic disorders
(such as renal disease),31 or medications (such as anticoagulants or β-blockers).32 A brief summary of common nail
changes associated with cutaneous disorders, nail cosmetics, and systemic disorders has been provided in Tables 2, ,3,3,
and and4,4, respectively.
Elderly patients might complain of common nail changes and dystrophies that cause pain, affect daily activities, are of
cosmetic concern, or are even malignant. Awareness of these conditions is essential for family practitioners, as well as
other specialists, to reach an accurate diagnosis and provide optimal management.
Nail changes are common among the elderly; however, they are often not brought to the attention of primary caregivers and
are thus overlooked. Nail changes among the elderly can either cause serious symptoms, impairing the daily activities of
this older population whose activities might already be restricted, or be asymptomatic but associated with substantial
cosmetic problems, leading to negative psychological effects. It is important that family physicians are knowledgeable about
and familiar with common nail abnormalities and their underlying causes in order to effectively reach an accurate diagnosis
and provide better care.

Levels of evidence
Level I: At least one properly conducted randomized controlled trial, systematic review, or meta-analysis
Level II: Other comparison trials, non-randomized, cohort, case-control, or epidemiologic studies, and preferably more than
one study
Level III: Expert opinion or consensus statements
This article has been peer reviewed.

Cet article a fait l’objet d’une révision par des pairs.


Both authors contributed to the literature search and preparation of the article for submission.
Competing interests

None declared /

Effect of fingernail length on the hand dexterity

Rikiya Shirato, OTR, PhD,1,* Atsumi Abe, OTR,2 Hikaru Tsuchiya, OTR,3 and Mizuki Honda, OTR4

Author information Article notes Copyright and License information Disclaimer

Dexterity is defined as appropriate voluntary activity used to manipulate objects during a specific task 1). Dexterity is among
the most important examination methods for the determination of neuromotor function of the hand, which involves
integration of motor and sensory functions. Dexterity also was the best predictor of independence in activities of daily
living2). Hand dexterity is affected by many factors, such as age, gender, educational level, and hand dominance3, 4).
Few previous studies, however, examined the effect of fingernail length on hand dexterity, and the few published studies
were conducted using artificial fingernails of several lengths attached to the subjects’ own fingernails 5, 6). Jansen et al.
reported that the degree of hand manipulation decreased with increased fingernail length, and they recommended that
patients cut their fingernails to 5 mm in length5). Uetake reported that the upper limit of fingernail length for handwork
efficiency and fashion considerations was 2 mm6).
Nandgaonkar also reported that the contribution of the fingernails to hand dexterity using the subjects’ own fingernails 7). It
was found that there is a significant correlation between the duration of fingernail cutting and the tasks on the modified Rivet
and Eyelet Deftness test7). However, there is a limitation in that the fingernail length was not quantified.
The fingernail consists of the nail plate, nail fold, nail bed and hyponychium. Beasley et al. suggested three basic categories
for the contributions of the fingernails to hand function; (1) fingernails are utilitarian, (2) they are environmental receptors for
sensory input, and (3) they are esthetic elements of our hands that are constantly exposed to scrutiny 8). Seah et al. also
reported that fingernails play a passive role in increasing sensory perception at the pulp by providing the counterpressure
necessary for the compression of sensory end organs between the palmar skin and the fingernail 9). Furthermore, Zook
reported that the fingernails allow for increased sensory perception in the pad of the finger as well as for the efficient and
accurate picking up of small objects10). The purpose of the present study was to clarify the effect of fingernail length on hand
dexterity using the subjects’ own fingernails as the fingernails play a role as perceptual receptors.


Thirty-eight healthy university student volunteers (eighteen males, twenty females) with a mean age of 21.4 years (range,
21–22 years) participated in the study. All subjects were right-hand dominant. None of subjects had a history of injury or
surgery to the upper extremities, or had neuromuscular diseases. The study was approved by research ethics review
committee of Hokkaido Bunkyo University (approval number: 28010), and informed consent was obtained from all subjects.
In order to evaluate fingernail size, the length and width of the nail on the right thumb, index finger and middle finger of each
subject were measured according to the method of Jung et al11). The length of the fingernail was defined as the greatest
longitudinal distance from the groove at the junction of the eponychium and proximal nail fold to the tip of the finger. The
width of the fingernail was defined as the greatest transverse distance between the two lowest points of the fingernail in the
lateral nail groove.
The length of the right thumb, index finger and middle finger, width of the respective distal interphalangeal joints, and length
and breadth of the right hand were also measured for each subject. The length of thumb and two fingers was defined as the
distance between the metacarpophalangeal joint and the fingertip. The distal interphalangeal joint (DIPJ) width was defined
as the distance across the DIPJ. Hand length also was defined as the distance from the distal wrist crease to the midpoint of
the tip of the middle finger. Hand breadth was also defined as the distance across the finger knuckles along the proximal
and distal palmar creases. These dimensions were measured using an electronic caliper (Shinwa Rules Co., Ltd., Sanjo,
The fingernail lengths to be compared in the study were set at 0 mm and 2 mm, which was regarded as both fashionable
and allowing efficient handwork based on the report of Uetake6). According to the previous study of Jansen et al.5), fingernail
length was defined as the length from tip of the finger to tip of the fingernail on the sagittal plane (Fig. 1). The state in which
the nail extended 0 mm from tip of finger was defined as a fingernail length of 0 mm, while that in which the nail extended 2
mm from tip of finger was defined as a fingernail length of 2 mm.
Fig. 1.
Measurement of fingernail length
The fingernail length was measured as the length from tip of the finger to tip of the nail on the sagittal plane. The fingernails
were cut to 0 mm or 2 mm of extension from tip of fingers.
Hand function and dexterity were assessed using the simple test for evaluating hand function (STEF, Sakai Medical Corp.,
Tokyo, Japan). The STEF was developed as a standardized battery by Kaneko et al. 12) for the simple and objective
evaluation of the functional movement ability of the upper limbs including the fingers. We selected this battery for use in the
study as the STEF was designed on basis of various hand activities12), and as it was used for the evaluation of motor skills
of the hand and upper extremities in a number of earlier studies13, 14).
The STEF battery consists of 10 types of subtests. Subjects were required to grasp, pinch or turn over objects of different
shapes and sizes and to carry them to an arranged area, and were valuate the speed of manipulation of objects using one
upper limb. The objects consisted of large sphere (70 mm in diameter, 5 pieces), middle-sized sphere (40 mm in diameter,
6 pieces), large rectangular box (100 × 100 × 47 mm, 5 pieces), middle-sized cube (35 × 35 × 35 mm, 6 pieces), thick
wooden disk (20 mm in diameter, 10 mm in thickness, 6 pieces), small cube (15 × 15 × 15 mm, 6 pieces), thin cloth (90 ×
80 mm, 6 pieces), thin metal disk (20 mm in diameter, 2 mm in thickness, 6 pieces), small sphere (5 mm in diameter, 6
pieces) and pin (3 mm in diameter, length in 42 mm, 6 pieces).
The time required for the manipulation of each object using the right hand measured in the study. First, STEF times were
measured after cutting the nails of the right thumb, index and middle fingers to 0 mm in length. As the fingernails grow at a
rate of about 3 mm a month15), after 20 days the fingernails were cut to 2 mm in length and STEF times were remeasured.
Measurements were taken in the sitting position three times at each fingernail length.
As the Shapiro-Wilks test indicated non-normal data distribution, non-parametric analyses were used. The differences
between males and females in term of age, and length and breadth of the right hand were compared using the Mann-
Whitney test. Finger length, DIPJ width and fingernail configuration of the right thumb, index finger and middle finger also
were compared using the Mann-Whitney test. The time of each STEF subtest with fingernails at 0 mm and 2 mm were
compared using the Wilcoxon signed-rank test. The times with fingernails at 0 mm and 2 mm were also compared between
males and females using the Wilcoxon signed-rank test. Furthermore, the times at each fingernail length were also
compared between males and females using the Mann-Whitney test.
All analyses were performed using SPSS Version 11.5 J for Windows. The level of significance was set at p<0.05.

The mean age was 21.5 years in males and 21.3 years in females. There was no significant difference between gender in
terms of mean age (p=0.20). The mean hand length and breadth were both significantly larger in males than in females
(p<0.01, p<0.01, respectively). The mean length of the thumb, index finger and middle finger were also larger in males than
in females (p<0.01, p<0.01, p=0.013, respectively). Moreover, the mean DIPJ widths of these fingers were similarly larger in
males (p<0.01, p<0.01, p<0.01, respectively). Although there were no significant differences in the mean fingernail length of
the thumb and index finger between genders (p=0.44, p=0.44, respectively), that of the middle finger was significantly larger
in males (p=0.03). The mean fingernail widths of the thumb, index finger and middle finger were also significantly larger in
males than in females (p<0.01, p<0.01, p<0.01, respectively) (Table 1).
Table 1.
General characteristics of the subjects
SD: standard deviation
aScores were compared between males and females using Mann-Whitney test.
*p<0.05, **p<0.01
The time for each STEF subtest with fingernails at 2 mm was generally shorter than that with fingernails at 0 mm. The
Wilcoxon signed-rank test showed that the mean times for the middle-sized sphere, small cube and pin with fingernails at 2
mm were significantly shorter than those with fingernails at 0 mm (p=0.03, p<0.01, p<0.01, respectively) (Table 2).
Table 2.
SD: standard deviation. aScores were compared between fingernails at 0 mm and 2 mm using Wilcoxon signed-rank test.
*p<0.05, **p<0.01
The Wilcoxon signed-rank test also showed that the mean times with fingernails at 2 mm were significantly shorter than
those with the fingernails at 0 mm in subtests with the thick wooden disk, small cube, and pin for males (p=0.03, p<0.01,
p<0.01, respectively). The mean times with fingernails at 2 mm were also significantly shorter than those with fingernails at
0 mm in subtests with the middle-sized cube, small cube, and pin for females (p=0.048, p<0.01, p<0.01, respectively) (Table
Table 3.
Times for STEF subtests with fingernails at 0 mm and 2 mm in males and females
SD: standard deviation. aScores were compared between fingernails at 0 mm and 2 mm in males using Wilcoxon signed-
rank test. bScores were compared between fingernails at 0 mm and 2 mm in females using Wilcoxon signed-rank
test. cScores were compared between males and females with fingernails at 2 mm using Mann-Whitney test. *p<0.05,
The Mann-Whitney test showed that the mean times with fingernails at 0 mm were not significantly different between males
and females (p>0.05, p>0.05, respectively), but the mean time with fingernails at 2 mm for females was significantly shorter
than that for males on the subtest with the small sphere alone (p=0.03) (Table 3).DISCUSSION
We observed that the time for each STEF subtest with fingernails at 2 mm was generally shorter than that with fingernails at
0 mm, and it was significantly shorter in a number of subtests. These findings indicated that fingernails at 2 mm are
advantageous to hand dexterity. In particular, fingernails cut to a length to allow them to be used to hook objects was
important for manipulation in the pin subtest, with fingernails cut at 2 mm appearing to have an advantage. However, in the
subtests with the middle-sized sphere, thick wooden disk and small cube, in which there is less need to hook objects with
the fingernails, fingernails of 2 mm in length also showed a significant effect. These results suggest that a certain length of
fingernail is necessary for performing dexterous manipulation in terms of the receptive function for sensory input proposed
by Beasley et al8).
Although the distribution of sensory receptors in the nail plate is not clear, mechanical forces applied to the nail plate are
transmitted to the mechanoreceptors in the nailbed, nail matrix and the nailfold 9). The distribution of Merkel cells, which are
slowly adapting type I mechanoreceptors, has been described in the nail matrix and nailfold 16, 17). Ruffini-like spray endings,
which are slowly adapting type II (SA-II) afferents, have also been reported in the nailbed18) and the half-moon19, 20).
Birznieks et al. have demonstrated that SA-II nail afferents, distributed in skin bordering the lateral edges of the nails,
respond reliably to forces applied to the fingertips that primarily come into contact with objects in manipulation tasks 21).
Furthermore, signals in populations of SA-II nail afferents contain directional information about fingertip forces.
The time taken for subtests in it was less necessary to hook the objects with the fingernails was significantly shorter with
fingernails at 2 mm in the present study. Fingernails at 2 mm in length might have advantages in transmitting information to
these mechanoreceptors distributed around the fingernail as the area of the fingernail is relatively larger than that for nails
cut to 0 mm.
Regarding the effect of fingernail length on hand performance, Jansen et al. investigated the speed and quality of finger
manipulations using a functional dexterity test with artificial fingernails extending 5 mm, 10 mm and 20 mm beyond the tips
of the fingers. It was revealed that the time and penalty increased as the length of the artificial fingernails increased. From
these results, they recommended that patients cut their fingernails to 5 mm in length5). Uetake examined the effect of
fingernail length on hand work efficiency with artificial fingernails of 4 different lengths: 0 mm, 2 mm, 4 mm and 6 mm. She
reported that work efficiency is best with fingernails at 0 mm, and work efficiency decreased as fingernail length increased,
so fingernail length should be limited to 2 mm or less for handwork efficiency and fashion considerations6).
In the present study, hand dexterity with fingernails at 2 mm was superior to that at 0mm, which is in disagreement with the
results of Uetake6). One possible reason could be that the sensory perception of fingertips differed from that in the Uetake
study6) as our study was performed using the subjects’ own fingernails. Nandgaonkar reported that the subjects’ own
fingernails definitely contribute to hand dexterity. As the number of days from when the fingernails were cut increased, the
dexterity score on the tasks requiring fingernails also increased7). However, fingernail length was not quantified.
The time required with fingernails at 2 mm for females was significantly shorter than that for males on the small sphere
subtest alone, and there were no significant gender-based differences in the other subtests in the present study. Previous
studies using Purdue pegboard tasks showed that females had advantage in hand performance compared to males 22, 23).
There is some reason to believe that the larger hand size of males handicaps their performance in the manipulation of small
objects23). In the present study, hand and fingers size were significantly smaller in females than in males. However, the
difference in hand performance between males and females was slight as the STEF used in this study was composed of 10
objects of various sizes.
The results from the present study indicate that the fingernail length of the subjects should be standardized when evaluating
hand dexterity and motor skill changes with time so that the hand dexterity is affected by fingernail length. We believe that
the influence of fingernail length on the evaluation of hand dexterity and motor skills could be eliminated by standardizing
fingernail length.
This study had a few limitations. First, subjects consisted of only healthy young adults. With increased age, the fingernails
flatten and the contours are modified11), resulting in platonychias and koilonychias24). Fine motor hand movements also
decline with aging25, 26), and the effect of the fingernail length on such movements in the elderly is unknown. Second, a
practice effect could have affected results for motor skills with fingernails at 2 mm despite the fact that the measurement
interval was more than 20 days. Further studies are necessary to clarify the influence of fingernail length on hand dexterity
in various ages.
In conclusion, the effect of the length of the fingernails on hand dexterity in subjects using their own fingernails was clarified
using the STEF. The time for each STEF subtest with fingernails at 2 mm was generally shorter than that with fingernails at
0 mm, and it was significantly shorter in number of subtests. There was little significant difference in the time taken for each
subtest between males and females. Our findings suggest that the fingernail length of subjects should be standardized
when evaluating the hand dexterity and motor skill changes with time.

We would like to offer special thanks to all the participants for their assistance on data collection.