Received: 5 October 2021 | Accepted: 7 December 2021

DOI: 10.1002/jmv.27514

SHOR T COMMUN I C A T I ON

Anon‐invasivemethod for diagnosingplantarwarts causedby
human papillomavirus (HPV)

Sara García‐Oreja1 | Francisco Javier Álvaro‐Afonso1 |

David Sevillano‐Fernández2 | Aroa Tardáguila‐García1 | Mateo López‐Moral1 |

José Luís Lázaro‐Martínez1

1Instituto de Investigación Sanitaria del

Hospital Clínico San Carlos (IdISSC), University

Podiatric Clinic, Complutense University of

Madrid, Madrid, Spain

2Microbiology Section, Medical Faculty of the

Complutense University of Madrid, Madrid,

Spain

Correspondence

Francisco Javier Álvaro‐Afonso, Universitary
Podiatric Clinic, Complutense University of

Madrid. Edificio Facultad de Medicina.

Pabellón 1. Avda. Complutense s/n, 28040,

Madrid, Spain.

Email: alvaro@ucm.es

Abstract

The methods that are used for the diagnostic confirmation of human papillomavirus

(HPV) include excisional biopsy and histopathological studies or polymerase chain

reaction (PCR). They are invasive, laborious, and subject to ethical restrictions due to

the benign nature of these warts. This study aims to analyse the accuracy of non-

invasive swab samples to diagnose plantar warts. Fifty plantar warts were included in

the study. Skin swabs and hyperkeratosis skin scales were collected from each wart.

Multiplex PCR was performed to detect and type the HPVs. The prevalence of HPV

in this study was 90% when the sample was obtained using the wart scraping

method and 94% when it was obtained using swabs and the new method. In 45 of

the 45 positive samples (sensitivity: 100%), the result between the wart scab and

wart swab were almost identical. The genotyping result was identical in all 46 pa-

tients who had a positive result using both methods. The swab method appears to be

a simple and accurate technique to diagnose plantar warts due to HPV. It is a

noninvasive technique that could be performed even by inexperienced professionals

and in patients with pain or a fear of needles.

K E YWORD S

cutaneous warts, HPV, human papillomavirus, PCR method, plantar warts

1 | INTRODUCTION

Skin warts are benign skin lesions that are caused by the human

papillomavirus (HPV), among which common warts and plantar warts

the most common.1,2 There are over 200 strains of the virus dis-

tributed over five genera, and 49 are based on their DNA sequences.3

HPV 2, 10, 27, and 57 from the alpha genus, HPV 4 from the gamma

genus, and HPV 1 from the mu genus are the most prevalent types

detected in plantar warts.2,4

An estimated 40% of the general population is infected with

HPV, of whom 7%–12% will develop a skin wart.4,5 The annual in-

cidence of plantar warts is 14%.5,6 Immunosuppression, having a wart

before, walking barefoot, and having a partner with a wart are the

main risk factors for developing a plantar wart.4,7,8

The most frequent clinical signs are the appearance of hy-

perkeratosis, which is due to the proliferation of mutated cells in the

foot, visualisation of thrombosed capillaries, pinpoint bleeding when

debriding hyperkeratotic tissue, and the disruption of normal

J Med Virol. 2021;1–5. wileyonlinelibrary.com/journal/jmv | 1

This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which permits use and distribution in any

medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made.

© 2021 The Authors. Journal of Medical Virology published by Wiley Periodicals LLC

http://orcid.org/0000-0002-2937-7122
http://orcid.org/0000-0003-4674-3822
http://orcid.org/0000-0001-5650-7351
http://orcid.org/0000-0001-7342-059X
http://orcid.org/0000-0001-6110-0265
mailto:alvaro@ucm.es
http://crossmark.crossref.org/dialog/?doi=10.1002%2Fjmv.27514&domain=pdf&date_stamp=2021-12-15


dermatoglyphics.6,7 In addition, pain on lateral compression of the

wart is common.9

The diagnosis is mainly clinical and must be confirmed via a

biopsy and histopathologic studies (gold standard) 2,9 using poly-

merase chain reaction (PCR) results from tissue samples or hy-

perkeratosis scale samples that were obtained from the wart, which

will also provide information on the virus biotype.1,9,10 Taking a

biopsy or scraping the hyperkeratosis to extract the DNA are labor-

ious procedures for clinicians, painful for the patient, and subject to

ethical restrictions because of the benign nature of these warts.1,9

Noninvasive swab samples have been used in two epidemiolo-

gical studies of HPV types of cutaneous warts by Bruggink et al.1,2 In

2011, De Koning et al.9 compared the results of HPV genotyping

from swab samples and wart scabs with those obtained from a biopsy

in 25 patients who were being treated for persistent cutaneous

warts. Swabs of the overlaying skin of each the wart were taken by

firmly rubbing a pre‐wetted cotton‐tipped stick five times over the

lesion surface.9 These authors found that wart scabs and swabs

showed identical viral typing results compared to the biopsies in 24

of 25 patients (sensitivity: 96%).9

To the best of our knowledge, no study has been published

comparing the swab method and the scraping method of wart

hyperkeratosis to diagnose plantar warts. This study aims to analyse

the accuracy of swabs for diagnosing plantar warts.

2 | MATERIALS AND METHODS

2.1 | Patients and samples

We prospectively collected data from all patients who attended the

University Podiatry Clinic of the Complutense University of Madrid

between December 2019 and December 2020. Patients older than

5 years of age who presented clinical signs of a plantar wart (hy-

perkeratosis, pinpoint bleeding when debriding hyperkeratotic tissue,

disruption of normal dermatoglyphics or pain on lateral compression

of the wart)6,7,9 (Figure 1A) were included in the study. Patients with

an allergy or hypersensitivity to cotton and patients with a wound or

bleeding in the wart area at the time of microbiological sampling were

excluded from the study.

Wart samples were obtained using swabs and scraping by a

single experienced researcher. After the wart was cleansed with

sterile saline, a sterile cotton swab was gently rotated five times over

the lesion surface (Figure 1B). The swab stick was placed into the

transport tube without preservation medium, and it was stored at

5–8°C until use. Later, the scab from the warts was obtained using a

scalpel by scraping on the hyperkeratotic surface of the lesion into an

Eppendorf tube (Figure 1C), and the sample was stored.

2.2 | HPV identification

All of the samples were processed within 24 h after collection.

Genomic DNA was extracted from hyperkeratosis scales and

swabs using the NZY Tissue gDNA Isolation kit (NZYTech), according

to the manufacturer's instructions, after 5 h of sample pre‐lysis at

56°C. For swab samples, the protocol was slightly modified. Before

pre‐lysis incubation, the swabs were resuspended in the pre‐lysis

solution (acid buffer NT1 and 25 μl Proteinase K) by vortexing for

1min at 3500 rpm. Subsequently, the pre‐lysis solution that was re-

tained on the cotton was drained by pressing the swab over the walls

of the tube.

DNA concentration was quantified after extraction at 260 nm

using a Nanodrop 2000 spectrophotometer. The purity was

determined by absorbance ratio 260 nm/280 nm.

HPV DNA amplification (one hundred ng of DNA) was performed

using a multiplex PCR reaction or by nested PCR (Mastercycler®

Nexus GSX1) using the NZYTaq II 2x Green Master Mix (NZYTech).

Multiplex PCR was performed for detecting and typing HPVs that are

related to verrucae vulgaris (HPV‐1, −2, −27, and −57) using the

specific primers and conditions that were proposed by Lei et al.11 The

nested PCR was performed using the primers and conditions that

were previously described by Forslund et al.12 for DNA samples that

were not amplified in the multiplex PCR followed by sequencing for

HPV typing. Positive and negative controls were included in every

amplification cycle. Amplification of the β‐globin gene using primers

PC04 and GH20 was used as internal control to assess the integrity

of human DNA. An amplicon of 286 bp was considered positive and

adequate the sample amount for HPV detection. All samples, in-

cluding HPV negative samples, resulted positive for β‐globin test. In

F IGURE 1 (A) Clinical signs of plantar warts. (B) Sampling using the swab method. (C) Sampling using the scraping method

2 | OREJA ET AL.



any case, DNA was extracted again, and the amplification was

repeated for HPV negative samples and samples with a nontyp-

able HPV.

The PCR products were purified using a NZYGelpure PCR pur-

ification kit (NZYTech) and sequenced with the Sanger method with a

BigDye terminator v3.1 cycle sequencing kit (Applied Biosystems)

using the nested PCR primers and an ABI 3730 XL genetic analyser

(Applied Biosystems). Sequences were analysed using Sequencing

Analysis software v.5.1 (Applied Biosystems) and the HPV

sequence alignment was determined through comparison with

known sequences in the GenBank database using BLASTn software

(https://blast.ncbi.nlm.nih.gov/).

2.3 | Statistical analyses

Statistical analysis was performed using SPSS v.22 (IBM Corp.). Fre-

quency and descriptive analyses were performed. We used the χ2

test to compare the qualitative variables and a Student's t‐test for the

association of qualitative variables with quantitative variables.

The sensitivity, specificity, positive and negative predictive va-

lues, and positive and negative likelihood ratios for the swab proce-

dure in diagnosing HPV were determined by comparing them with

the HPV results that were generated from the wart scabs. For this,

we used Epidat v.3.1 (Galicia, Spain).

The sample size was estimated to be 46 samples with a statistical

power of 0.80 and an alpha of 0.05, using the GRANMO Sample Size

Calculator version 7.12 Online (Institut Municipal d'Investigació

Mèdica, Barcelona, Spain).

3 | RESULTS

A total of 50 samples from 46 patients were analysed. Four patients

had 2 warts and in these cases 2 samples were taken and analysed,

one for each wart. The mean age of the people studied was

49.46 ± 20.41 years with a normal distribution. Eighteen participants

(36%) were men and 32 (64%) were women. Twenty‐seven (54%)

lesions were in the plantar area of the forefoot, 16 (32%) were on the

heel, six (12%) were in the ball of the fingers, and one (2%) was

periungual. Thirty‐six (72%) lesions were considered to be re-

calcitrant because they had an evolution time greater than

6 months,13 and 14 (28%) were considered to be non‐recalcitrant.

The most frequent clinical signs were as follows: 48 (96%) warts

presented hyperkeratosis, 47 (94%) showed disruption of normal

dermatoglyphics, 29 (58%) had pinpoint bleeding, and 48 (96%) had

pain on lateral compression.

The prevalence of HPV in this study was 90% when the sample

was obtained using the wart scraping method and 94% when it was

obtained using the new swab method (Table 1). Discordant results

were obtained between methods for only two samples (Table 1). In

these two samples who had positive swab results, the scraping results

were negative.

One of the negative swabs was taken from a wart located on the

plantar area of the forefoot and the other 2 negative swabs were

taken from warts located on the heel (p = 0.594). The three swabs

with a negative result were taken from recalcitrant lesions (p = 0.550)

and one was taken from a lesion that had been previously treated

with a keratolytic (p = 0.424).

The genotyping results were identical in the 45 samples, for

whom both methods had a positive result. The most prevalent HPV

biotype was 2, which was present in 26.7% (12/45) of samples. Seven

samples were positive for HPV57 (15.6%), seven for HPV19 (16.6%),

six for HPV5 (13.3%), four for HPV27 (8.9%), three for HPV1 (6.7%),

two for HPV20 (4.4%), one for HPV14 (2.2%), one for HPV50 (2.2%),

one for HPV65 (2.2%), and one was nontypeable HPV (2.2%). In the

two samples in whom there was a discordant result, swab genotyping

showed HPV65. No sample had a result of more than one biotype in

the multiplex PCR. DNA was extracted again, and the amplification

was repeated for the sample with a non‐typeable HPV, and the

results were identical.

Diagnostic accuracy data (with 95% confidence intervals) for the

swab versus the scraping method were as follows: sensitivity of 1

(0.99–1), specificity of 0.6 (0.7–1), positive predictive value of 0.96

(0.89–1), negative predictive value of 1 (0.83–1) and positive

likelihood ratio of 2.5 (0.85–7.21).

4 | DISCUSSION

In this study, we observed that in 45 of the 45 positive samples

(sensitivity: 100%), the result between wart scab and the wart swab

were identical. Therefore, using swabs is a sensitive test to con-

firmation an HPV diagnosis, and it has the advantage of being less

invasive for the patient as well as simpler, easier, and faster to per-

form for the clinicians. De Koning et al.9 observed a similar sensitivity

(96%) for swabs and scraping compared to the gold standard test

(biopsy) in a sample of 25 patients. However, we observed that the

swab method has a specificity of 60% compared with scraping

method. Previous studies have not calculated the specificity of swabs

because they did not include lesions that were not clinically diag-

nosed as skin warts.9 In our study, only patients with clinical signs of

a plantar wart were included, which could lead to a bias when

TABLE 1 HPV prevalence data and PCR results by sampling
method

Wart scab
HPV negative HPV positive Total

Swabs

HPV negative 3 0 3 (6%)

HPV positive 2 45 47 (94%)

Total 5 (10%) 45 (90%) 50

Abbreviation: HPV, human papillomavirus; PCR, polymerase chain
reaction.

OREJA ET AL. | 3

https://blast.ncbi.nlm.nih.gov/


calculating the specificity. In the present study, the swab method

showed a positive predictive value of 0.96, a negative predictive

value of 1, and a positive likelihood ratio of 2.5 compared with

scraping method. The positive predictive value of 0.96 for our study

population is not surprising because our HPV prevalence was 90%

when the sample was obtained using the wart scraping method

(which we consider to be the gold standard of noninvasive methods).

De Koning et al.9 had a prevalence of 96% in their study using a

biopsy and a similar prevalence between scraping and swab methods

(92%). A randomized controlled trial that was conducted by Bruggink

et al.1 used the swab method and PCR as a diagnostic method. They

reported only 7% negatives in a sample of 250 participants with 391

common warts and 379 plantar warts. In the present study, HPV

prevalence was also 94% using the swab method. To the best of our

knowledge, other diagnostic precision calculations, such as positive

and negative predictive values and positive and negative likelihood

ratios, have not been reported in previous studies.

There was no statistically significant association between nega-

tive swab results and wart location, recalcitrant or nonrecalcitrant

wart and whether the patient had previously received treatment,

which in our opinion may be due to the lower number of negative

cases.

For genotyping, all positive results were identical between the

sampling methods. Genotyping results that were obtained from scabs

and swabs of the warts in the study by De Koning et al.9 were

identical to the results from the deeper wart portions (wart biopsy) in

24 of the 25 (96%) patients. In two patients, the result between the

wart biopsy and the wart scab or wart swab were not identical, with

the scab and wart swab showing negative results.9 Both sample

collection methods yielded identical results in the sample where HPV

genotyping was not possible (HPV nontypeable). In our opinion, as-

suming that the amount of DNA was optimal, the nontypeable gen-

otypes could indicate an overlapping sequence of more than one

HPV in this sample.

In the present study there were two samples in whom the swab

had a positive result, but the results of scraping (which is considered

to be the gold standard of noninvasive tests) were negative. Both

patients had a HPV65 genotype, which is a type of Gammapapillo-

mavirus (GammaPV) that is frequently isolated in common and

plantar warts.10,14–16 De Koning et al.17 recently showed the pre-

sence of HPV in up to 80% of swab samples of clinically normal skin,

thus raising the question of whether HPV types detected on wart

surfaces represent colonization or a latent or productive HPV in-

fection.16 We consider that the patients who had a positive swab

result had an active HPV infection because these patients had clinical

signs of a plantar wart (hyperkeratosis, pinpoint bleeding when

debriding hyperkeratotic tissue, disruption of normal derma-

toglyphics, or pain on lateral compression of the wart).6,7,9

This study has some limitations. The results of the swab method

were not compared with the biopsy, which is considered to be the

gold standard for diagnosing HPV. However, this test is considered to

be a very invasive test. Additionally, patients without clinical signs of

a wart were not included, which could lead to a bias in calculating the

specificity. Future studies could use the swab method in patients with

healthy skin, but a comparison with the biopsy or scraping would be

difficult because they are invasive methods for the patient. On the

other hand, in future studies it would be advisable to sample the

same wart by different practitioners to evaluate the interobserver

realiability of this test.

In conclusion, the swab method appears to be a simple and accu-

rate technique to diagnose plantar warts that are caused by HPV. It is a

noninvasive technique that could be performed by even inexperienced

professionals and in patients with pain or a fear of needles.

FUNDING INFORMATION

Funding information is not available.

CONFLICT OF INTERESTS

The authors declare that there are no conflict of interests.

AUTHOR CONTRIBUTIONS

Sara García Oreja, Francisco Javier Álvaro Afonso, and José Luís

Lázaro Martínez: had full access to all data in the study and were

responsible for their integrity. Sara García Oreja: provided the study

concept and design. Sara García Oreja and Francisco Javier Álvaro

Afonso: obtained the data. David Sevillano‐Fernández: processed

and analysed the samples in the microbiology laboratory. Aroa

Tardáguila‐García and Mateo López‐Moral: statistically analysed the

data. All authors and interpreted the data, drafted the manuscript,

and critically revised the manuscript.

ETHICS AND CONSENT STATEMENT

The protocol was approved by the medical ethics committee of the

Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC)

(protocol number 20/013‐E). All patients voluntarily signed consent

statement for the use of their image and publication of their case details.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the

corresponding author upon reasonable request.

ORCID

Sara García‐Oreja http://orcid.org/0000-0002-2937-7122

Francisco Javier Álvaro‐Afonso http://orcid.org/0000-0003-

4674-3822

Aroa Tardáguila‐García http://orcid.org/0000-0001-5650-7351

Mateo López‐Moral http://orcid.org/0000-0001-7342-059X

José Luís Lázaro‐Martínez http://orcid.org/0000-0001-6110-0265

REFERENCES

1. Bruggink SC, Gussekloo J, de Koning MN, et al. HPV type in plantar
warts influences natural course and treatment response: secondary
analysis of a randomised controlled trial. J Clin Virol. 2013;57(3):227‐232.

2. Bruggink SC, de Koning MN, Gussekloo J, et al. Cutaneous wart‐
associated HPV types: prevalence and relation with patient char-
acteristics. J Clin Virol. 2012;55(3):250‐255. https://pubmed.ncbi.
nlm.nih.gov/22884670/

4 | OREJA ET AL.

http://orcid.org/0000-0002-2937-7122
http://orcid.org/0000-0003-4674-3822
http://orcid.org/0000-0003-4674-3822
http://orcid.org/0000-0001-5650-7351
http://orcid.org/0000-0001-7342-059X
http://orcid.org/0000-0001-6110-0265
https://pubmed.ncbi.nlm.nih.gov/22884670/
https://pubmed.ncbi.nlm.nih.gov/22884670/


3. Bzhalava D, Eklund C, Dillner J. International standardization and
classification of human papillomavirus types. Virology. 2015476:
341‐344.

4. García‐Oreja S, Álvaro‐Afonso FJ, García‐Álvarez Y, García‐Morales

E, Sanz‐Corbalán I, Lázaro Martínez JL. Topical treatment for plantar
warts: a systematic review. Dermatol Ther. 2021;34(1):e14621.

5. Witchey D, Witchey N, Roth‐kauffman M, Kauffman M. Plantar
warts: epidemiology, pathophysiology, and clinical management.
J Am Osteopath Assoc. 2018;118(2):92‐105.

6. Hashmi F, Bristow I. Treating plantar warts: utilising natural im-
munity to induce wart regression. Dermatological Nursing. 2014;
13(1):42‐46.

7. Vlahovic TC, Khan MT. The human papillomavirus and its role in
plantar warts: a comprehensive review of diagnosis and manage-

ment. Clin Podiatr Med Surg. 2016;33(3):337‐353. doi:10.1016/j.
cpm.2016.02.003

8. Gibbs S, Harvey I. Topical treatments for cutaneous warts (Review).
Cochrane Database Syst Rev. 2006;(3)):001781.

9. de Koning MN, Khoe Lv, Eekhof JA, et al. Lesional HPV types of

cutaneous warts can be reliably identified by surface swabs. J Clin Virol.
2011;52(2):84‐87. https://pubmed.ncbi.nlm.nih.gov/21798797/

10. de Koning MN, ter Schegget J, Eekhof JA, et al. Evaluation of a novel
broad‐spectrum PCR‐multiplex genotyping assay for identification

of cutaneous wart‐associated human papillomavirus types. J Clin

Microbiol. 2010;48(5):1706‐1711.
11. Lei YJ, Gao C, An R, et al. Development of a multiplex PCR method

for detecting and typing human papillomaviruses in verrucae vul-
garis. J Virol Methods. 2008;147(1):72‐77.

12. Forslund O, Ly H, Higgins G. Improved detection of cutaneous
human papillomavirus DNA by single tube nested “hanging droplet”
PCR. J Virol Methods. 2003;110(2):129‐136. https://pubmed.ncbi.
nlm.nih.gov/12798239/

13. Kaçar N, Taşli L, Korkmaz S, Ergin Ş, Erdoǧan BŞ. Cantharidin‐
podophylotoxin‐salicylic acid versus cryotherapy in the treatment of

plantar warts: a randomized prospective study. J Eur Acad Dermatol

Venereol. 2012;26(7):889‐893.

14. Egawa N, Egawa K, Griffin H, Doorbar J. Human papillomaviruses;
epithelial tropisms, and the development of neoplasia. Viruses. 2015;

7:3863‐3890.

15. Lee HS, Lee JH, Choo JY, Byun HJ, Jun JH, Lee JY. Im-

munohistochemistry and polymerase chain reaction for detection
human papilloma virus in warts: a comparative study. Ann Dermatol.
2016;28(4):479‐485. doi:10.5021/ad.2016.28.4.479

16. Breznik V, Fujs Komloš K, Hošnjak L, et al. Determination of
causative human papillomavirus type in tissue specimens of
common warts based on estimated viral loads. frontiers in cellular

and infection. Microbiology. 2020;10:4. doi:10.3389/fcimb.2020.
00004/full

17. de Koning MNC, Quint KD, Bruggink SC, et al. High prevalence of
cutaneous warts in elementary school children and the ubiquitous
presence of wart‐associated human papillomavirus on clinically
normal skin. Br J Dermatol. 2015 172(1), 196‐201. doi:10.1111/bjd.
13216

How to cite this article: García‐Oreja S, Álvaro‐Afonso FJ,

Sevillano‐Fernández D, Tardáguila‐García A, López‐Moral M,

Lázaro‐Martínez JL. A non‐invasive method for diagnosing

plantar warts caused by human papillomavirus (HPV). J Med

Virol. 2021;1‐5. doi:10.1002/jmv.27514

OREJA ET AL. | 5

https://doi.org/10.1016/j.cpm.2016.02.003
https://doi.org/10.1016/j.cpm.2016.02.003
https://pubmed.ncbi.nlm.nih.gov/21798797/
https://pubmed.ncbi.nlm.nih.gov/12798239/
https://pubmed.ncbi.nlm.nih.gov/12798239/
https://doi.org/10.5021/ad.2016.28.4.479
https://doi.org/10.3389/fcimb.2020.00004/full
https://doi.org/10.3389/fcimb.2020.00004/full
https://doi.org/10.1111/bjd.13216
https://doi.org/10.1111/bjd.13216
https://doi.org/10.1002/jmv.27514