SEARCH

SEARCH BY CITATION

Keywords:

  • dermatophytes;
  • PCR;
  • diagnostics;
  • quality standard;
  • onychomycosis;
  • tinea pedis;
  • tinea capitis

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. False-negative results
  5. PCR formats
  6. Future directions
  7. Conflict of interest
  8. References

The prevalence of onychomycosis is increasing steadily, sevenfold alone in the US within the last twenty years. An important aspect in this development is the demographic development of the human population of the industrial countries like Germany. A fast and accurate laboratory diagnosis is essential for successful treatment because 50% of the cases are misdiagnosed when relying on the clinical appearance only. The current diagnosis of dermatophytosis, based on direct microscopy and culture of the clinical specimen, is problematic given the lacking specificity of the former and the length of time needed for the latter. Molecular techniques can help to solve these problems. In recent years, a number of in-house PCR assays have been developed to identify dermatophytes directly from clinical specimens. Based on the “Mikrobiologisch-infektiologischen Qualitätsstandards (MIQ) für Nukleinsäure-Amplifikationstechniken” and the MIQE guideline (Minimum Information for Publication of Quantitative Real-Time PCR Experiments) 11 studies are reviewed which were published between 2007 and 2010. The present article evaluates the quality of the PCR assays regarding false positive and false negative results due to contamination, PCR format, statistical analysis, and diagnostic performance of the studies. It shows that we are only at the beginning of providing high quality PCR diagnosis of dermatophytes.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. False-negative results
  5. PCR formats
  6. Future directions
  7. Conflict of interest
  8. References

In recent years the prevalence of onychomycosis has been steadily increasing. In the United States, for instance, the prevalence increased from 2.18% in 1979 to 13.8% in 2000 [1, 2]. The “Achilles Project”, a 16-country European study that surveyed patients who had visited a general practitioner, reported a prevalence of 26%[3].

There are a number of predisposing factors that may increase the risk of fungal infection of the foot. These include a genetic disposition and underlying diseases such as diabetes and immune deficiency as well as participation in sports, exposure to environmental factors, and demographic developments [4]. In industrialized nations, about 20% of the population over 60 years of age and 50% of the population over age 70 has a fungal infection of the foot [5, 6]. In 20 years this segment of the elderly population will make up 15–35%[7]. Over the course of a single year (1989–90), the treatment of onychomycosis in the United States cost the healthcare system 43 million dollars [8]. In 1999, 250 million US dollars were spent on debridement of mycotic nails alone [9]. It is too soon to predict just how high the costs will be in 20 years there or in Germany.

For successful treatment of onychomycosis, as well as any dermatophytosis, accurate and rapid diagnosis is essential. If the diagnosis is based on clinical symptoms alone, about 50% of patients are presumably misdiagnosed [10]. Laboratory diagnosis is based on direct microscopic analysis and cultivation using collected clinical material. Significant disadvantages include the lacking specificity of microscopy and the time needed to grow a fungal culture (4–6 weeks). In addition, 30–50% of microscopically identified cases of onychomycosis cannot be cultivated and hence the pathogen cannot be identified at the species level. One reason is that due to the commercial availability of topical antifungal agents, many patients have already treated the affected area themselves, impairing proper cultivation.

This contributes to the fact that 40% of dermatologists in Germany [11] reportedly initiate therapy without actually identifying the cause of the nail disorder. From a medical, economic, and legal standpoint, the use – especially of systemic antifungal agents – is unwarranted under such circumstances, particularly given the potential for side effects and drug interactions [12–14].

With the advent of molecular diagnosis as a microbiological detection method, we now have a tool at our disposal for accurate and rapid diagnosis of dermatophytosis (even at the species level) within 48 hours. “In-house” PCRs are already available and reports on their use have been published (Table 1). A test kit (Dermatophytes – Multiplex by SSI Diagnostica, Denmark), which can only identify T. rubrum at the species level as well as dermatophytes as a group (including geophilic species, which are often only contaminants), is commercially available. Another test kit, by the Dresden-based company, Biotype (Mentype – ycoderm), has been available since the beginning of the year. Yet this raises the question which test kit or which “in-house” PCR is most suitable for delivering highly specific and sensitive results? One criterion for making the right decision are the quality standards for molecular diagnosis which appeared in the brochure “Mikrobiologisch-infektiologische Qualitätsstandards (MIQ)” for nucleic acid amplification techniques which was revised in 2011 [15] or in the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines [16], the latter of which should not be restricted to real-time (RT)-PCRs. In the following we discuss some of the most important points from the guidelines concerning the establishment and use of molecular diagnosis [17]. We selected and analyzed 11 studies published between 2007 and 2010 on the development of dermatophyte PCR (Table 1). Our main focus was on studies that had tested the method on clinical material and which had used a commercially-available test kit for DNA extraction.

Table 1.  Studies included in the analysis of quality standards.
  1. CHS = chitinase, ITS = internal transcribed spacer of ribosomal DNA, LSU = large subunit of rDNA, IAC = internal amplification control, EC = extraction control EP = endpoint, RT = realtime.

Reference Target gene Controls PCR format Target species DNA extraction method Analytic sensitivity Analytic specificity -not present PV NV
Litz 2010CHS1IACEP, nestedAll dermatophyte spp.Freeze-thaw alk. lysis1-5 cells Chrysosporium spp.
Brillowska 2010ITSEPAll trichophyton spp. M. canis, M. audouiniiAlk. lysis BSA (antiinh) M. canis M. ferrugineum
Uchida 2009ITS1EP, nested T. rubrum T. interdigitale Mechan. lysis, phenol/chloroform T. violaceum T. interdigitale (zoophil)
Ebihara 2009LSUEP, nested T. rubrum T. interdigitale Mechan. lysis phenol/chloroform100–1000 cells T. interdigitale (zoophil)
Bergmans 2009ITSEC IACRT11 dermatophyte spp.Lysis (Prot K), Mag NA Pure1 pg 
Bergmans 2008ITSEC IACEP9 dermatophyte spp.Lysis (Prot K) QuiAmp kit0.1–1 pg 
Garg 2007CHS1EP, nestedAll dermatophyte spp.Lysis (Prot K), phenol/chloroform Chrysosporium spp.
Li 2007ITSEP17 dermatophyte spp.Mechan. lysis, kit1–10 pg T. tonsurans
Arabatzis 2007ITSEC IACRT6 Spp.Lysis (Prot K) QuiAmp kit0.1 pg M canis/audouinii T. schönleinii 95.7100
Brillowska 2007CHS ITSIACEP, multiplexAll dermatophyte spp. T. rubrumAlk. lysis BSA (antiinh) Chrysosporium spp.
Savin 2007CHSEP, broad rangeFungi, All dermatophyte spp.High pure template prep kit1 fg Chrysosporium spp.

Technical requirements

Separate rooms should be available for all necessary steps in the procedure, from pre- to post-PCR. Aerosol-resistant pipette tips are used in nearly every step. Certain procedures, such as DNA extraction or starting the PCR reaction, are done using microbiological safety benches. Only one of the 11 studies reported on this.

False-positive results

In order to avoid false-positive results from contamination during DNA extraction one must:

  • • 
    ensure that only tested fungus-free reagents are used for DNA extraction and PCR setup and that
  • • 
    10% of specimens should be fungus-free extraction controls collected from the same tissue (nails, hair, skin) as the clinical material being tested.
    Only about one-fourth of studies report on these two points.
    In order to avoid false-positive results due to contamination during PCR setup,
  • • 
    The UDG enzyme (uracil-DNA glycosylase) should be used. This requires that dUTP (instead of dTTP) is incorporated during the PCR into the newly synthesized DNA. PCR products generated in this manner may be destroyed by the enzyme after analysis and are thus not an important source of contamination.
    One quarter of studies used this option.
  • • 
    Especially when enhancing the detection threshold by using “nested” PCR, which involves accumulation of the initial PCR product in a second, independent PCR reaction by opening the first product and pipetting the first amplification product over it, one in ten PCR reactions should be a negative control (ddH2O instead of DNA).
    Out of 4 studies that used a “nested” PCR technique, none reported on this precautionary measure.
  • • 
    Cross-hybridization of primers, probes, and human DNA (from clinical material) should be ruled out. At the very least, one must show whether – and if so, then which – phylogenetically closely related species, or other species which are possible differentials, were also detected or could not be ruled out.
    Only about 25% of studies reported on this.

False-negative results

  1. Top of page
  2. Summary
  3. Introduction
  4. False-negative results
  5. PCR formats
  6. Future directions
  7. Conflict of interest
  8. References

Steps should also be taken to avoid false-negative results. These are largely due to the use of suboptimal methods for DNA extraction. Not only should the yield of various methods (mainly kits) be compared with one another, but also the volume used in the PCR reaction, which is often between 1–20 μl, as well as minimum and maximum amounts of processed clinical material.

An ideal DNA extraction method would be selectively accumulation of fungal DNA with a simultaneous reduction of human DNA. Given that such a method is not yet available, the greater the volume, the more fungal DNA is also available for amplification and thus the more sensitive is its detection. Yet in the worst case, the availability of more human DNA (from clinical material) for amplification can lead to cross-hybridization or inhibition of the PCR. Ensuring optimal conditions for sensitivity requires optimization of the above-mentioned parameters. None of the studies we evaluated reported doing so.

False-negative results may be generated during PCR by excessively high “foreign” DNA concentrations or inhibitors which are not removed during DNA extraction. Thus “internal amplification controls” (IAC) are also needed. In most cases (as in the 11 studies) these consist of either the species to be detected from the DNA (reference strain), which is amplified in a second reaction along with the tested specimen, or in the amplification of human DNA (from clinical material) with a second human-specific primer pair. The second option is less optimal, given that excessive amounts of human DNA are generally amplified, even with inhibitors, while the fungal DNA, which is present in smaller amounts, is already inhibited. Ideally, the amplification of IAC should be in the same specimen as the DNA from the patient specimen, but using a different primer pair (to avoid competition). The DNA is from other target regions (phage or artificial DNA sequence).

The analytical sensitivity of an assay ( = lowest number of copies in the specimen which may be accurately detected by testing) should not only be used for the PCR portion, but also should be included for the DNA extraction method. Most studies, however, only reported on amplification.

A final point is that when using “broad range” or “multiplex” PCRs, which use either universal primers that can bind to the DNA of several species or species-specific primer pairs in a reaction, which can yield various fungal species simultaneously, it is essential to detect to what extent a mixed infection can be identified or where the detection limit for the pathogen (found in lesser amounts) is.

None of the studies reported specifically on this procedure.

PCR formats

  1. Top of page
  2. Summary
  3. Introduction
  4. False-negative results
  5. PCR formats
  6. Future directions
  7. Conflict of interest
  8. References

A general distinction is made between endpoint PCR (EP) and (quantitative) RT-PCR. Most PCRs for the detection of dermatophytes are conventional PCRs using endpoint analysis, i.e., the result is apparent only at the end of the reaction and thus one cannot follow the accumulation of the product as it takes place. Quantification of the product or conclusions about the amount of starting material as in real-time PCR are not possible. Only 2 of the published studies concerned real-time PCR assays, which have the advantages of being faster and less susceptible to contamination. Species detection may be performed, for instance, by using labelled probes contained within the same reaction tube. Conventional PCR, on the other hand, performs detection externally, after amplification by means of hybridization (ELISA, Western blotting, etc.) or sequencing. Such methods are not only more costly in terms of time and personnel, but due to the necessary manipulations are also more susceptible to contamination. They have the advantage, however, of lower cost (equipment and materials).

Statistical analysis and diagnostic performance

The diagnostic sensitivity ( = percentage of people with a certain disease who test positive under certain conditions) and specificity ( = percentage of people without the disease who test negative under the conditions) of a test must be analyzed and should be around 95% probability for both options. In addition, the positive and negative predictive values (PV and NV) must be calculated. The PV says how many people who have been identified as having the disease by a given test actually had it. The NV says how many people were not positive according to the test and were indeed healthy. Only one of the 11 studies calculated the PV and NV. It should be noted, however, that calculating these parameters for dermatophyte PCR is difficult and may lead to confusing results. If, for instance, in onychomycosis, the results of culture (the current gold standard) are used as the basis for calculating the percentage of “true positives” and the “false-positive” percentage (i.e., negative culture results and positive PCR), this results in low values for diagnostic specificity and positive predictive value. The reason is that dermatophytes from nail material cannot be cultivated in up to 50%, although microscopic fungal filaments are visible and detectable on PCR. Yet using the proportion of microscopically and culturally positive results as a gold standard for calculations is warranted because microscopic analysis is too unspecific.

The dilemma about replacing the less sensitive and specific gold standard with a more sensitive and specific method applies to a number of pathogens. A realistic, basic solution is needed.

Future directions

  1. Top of page
  2. Summary
  3. Introduction
  4. False-negative results
  5. PCR formats
  6. Future directions
  7. Conflict of interest
  8. References

In our work as a national consultant laboratory for dermatophytes, we have developed a high-quality diagnostic procedure (real-time PCR) for dermatophytosis. This technique is based on amplification of a genome fragment, the internal transcribed spacer (ITS) region of ribosomal DNA of fungi and subsequent detection by hybridization using a species-specific nucleotide probe. Using a type of modular approach up to 9 species may be detected (Trichophyton tonsurans/T. equinum, T. interdigitale, T. schoenleinii/ T. mentagrophytes, T. verrucosum, Trichophyton species from Arthroderma benhamiae/T. concentricum/T. erinacei, T. rubrum, T. violaceum, E. floccosum, M. canis/M. ferrugineum/M. audouinii). This form of diagnosis was developed by the authors and is now available to all interested colleagues and patients. The pathogen is identified directly in clinical material. Analysis may be done with simple nail specimens, hair samples, and skin scrapings (minimum 3 mg). Specimens may be sent in by mail. Correct collection of material is essential. According to the fee schedule for physicians (GOÄ, from 09/2008), conventional dermatophytosis diagnosis (plain specimen, 2 cultures with and without antibiotic, microscopic identification of the cultivated fungus) costs 26.80 € and molecular biological analysis (DNA extraction, amplification, species-specific probe detection) costs 99.09 € (basic fee). The higher costs for acquiring a real-time PCR machine compared with light microscopy must also be taken into account. Thus this type of diagnosis is currently only feasible for microbiological laboratories which also identify other pathogens (bacteria, viruses, parasites) with molecular genetic methods. Culture and plain specimen must still be done in parallel for all patients in which specific pathogen detection for the most common dermatophyte species related to the respective clinical appearance is negative (for instance, for T. rubrum, T. interdigitale and E. floccosum in suspected onychomycosis/tinea pedis) (Figure 1). Rare dermatophyte species, such as yeasts and molds are not detected by specific diagnosis; for determining resistances (essential for molds!) as well, cultivation is essential. An assay, which uses specific primers/probes to detect all anthropophilic/zoophilic and clinically relevant geophilic dermatophyte species (Trichophyton, Microsporum and Epidermophyton) as a group but not the other, non-relevant, geophilic dermatophyte species (contaminants), is desirable, but would fail given the close phylogenetic relationship of these fungal species.

image

Figure 1. Diagnostic strategy when fungal agents are suspected. Molecular detection of species in brackets when indicated only.

Download figure to PowerPoint

The advantages of performing a sensitive, correct, and rapid diagnosis of the most common dermatophytes at the species level are clear:

  • 1
    Detection and prompt initiation of adequate antifungal therapy (within the group of dermatophytes, for instance, Microsporum and Trichophyton species as well as zoophilic and anthropophilic species are treated differently).
  • 2
    For zoophilic pathogens, the transmitting animal (cat, rodent, rabbit, cow, horse, etc.) should be identified and also treated given that animal sources of infection are often asymptomatic.
  • 3
    Epidemiological considerations.

When updating/revising the guidelines (tinea affecting the skin, tinea capitis, onychomycosis), the use of molecular diagnostic techniques in fungal skin infection should be urgently considered.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. False-negative results
  5. PCR formats
  6. Future directions
  7. Conflict of interest
  8. References
  • 1
    Editorial. Prevalence, morbidity, and cost of dermatological diseases. Journal of Investigative Dermatology 1979; 73: 395401.
  • 2
    Ghannoum MA, Hajjeh RA, Scher R, Konnikov N, Gupta AK, Summerbell R, Sullivan S, Daniel R, Krusinski P, Fleckman P, Rich P, Odom R, Aly R, Pariser D, Zaiac M, Rebell G, Lesher J, Gerlach B, Ponce-de-Leon GF Ghannoum A, Warner J, Isham N, Elewski B. A large-scale North American study of fungal isolates from nails: the frequency of onychomycosis, fungal distribution, and antifungal susceptibility patterns. Journal of the American Academy of Dermatology 2000; 43: 641648.
  • 3
    Haneke E, Roseeuw D. The scope of onychomycosis: epidemiology and clinical features. International Journal of Dermatology 1999; 38: 712.
  • 4
    J. Thomas, G. A. Jacobson, C. K. Narkowicz, G. M. Peterson, H. Burnet, Sharpe C. Toenail onychomycosis: an important global disease burden. Journal of Clinical Pharmacy and Therapeutics 2010; 35: 497519.
  • 5
    Loo DS. Onychomycosis in the elderly: drug treatment options. Drugs and Aging 2007; 24: 293302.
  • 6
    Salgo PL, Daniel CR, Gupta AK, Mozena JD, Joseph SW. Onychomycosis disease management. Medical Crossfire: debates, peer exchange and insights in medicine 2003; 4: 117.
  • 7
    Hayutin AM. How population aging differs across countries: a briefing on global demographics. Stanford Centre on longevity 2007. Available at: http://longevity.stanford.edu/files/BriefingGlobal
  • 8
    Elewski BE. Onychomycosis: treatment, quality of life, and economic issues. American Journal of Clinical Dermatology 2000; 1: 1926.
  • 9
    Ele Tan JS, Joseph WS. Common fungal infections of the feet in patients with diabetes mellitus. Drugs and Aging 2004; 21: 101112.
  • 10
    Faergemann J, Baran R. Epidemiology. Clinical presentation and diagnosis of onychomycosis. Br J Dermatol 2003; 149: Suppl 65:14.
  • 11
    Effendy I, Lecha M, Feuilhade de Chauvin M., Di Chiacchio N, Baran R. European Onychomycosis Observatory. Epidemiology and clinical classification of onychomycosis. J Eur Acad Dermatol Venereol 2005; 19: Suppl 1: 812.
  • 12
    Agathos M. Diagnose und Behandlung von Mykosen. Serie diabetisches Fußsyndrom, Folge 6, MMW-Fortschr Med 2000; Nr. 44.
  • 13
    Ralph DC. 3rd and Elewski BE. The diagnosis of nail fungus infection revisited. Arch Dermatol 2000; 136, 11621164.
  • 14
    Shemer A, Trau H, Davidovici B, Grunwald MH, Amichai B. Nagelproben bei Onychomykose: Vergleichsstudie mit einer Kurettage an drei Stellen des infizierten Nagels. J Dtsch Dermatol Ges 2007; 5: 11081111.
  • 15
    Reischl U, Drosten C, Geißdörfer W, Göbel U, Hoffmann KS, Mauch H, Meyer T, Moter A, von Müller L, Panning M, Rabenau HF, Reiter-Ogawa I, Roth A, Weitz M. Urban & Fischer Verlag in Elsivier 2011; Heft 1, 3. Aufl.
  • 16
    Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, Mueller R, Nolan T, Pfaffl MW, Shipley GL, Vandesompele J, Wittwer CT. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clinical Chemistery 2009; 55: 611622.
  • 17
    Khot PD, Fredericks DN. PCR-based diagnosis of human fungal infections. Expert Rev Anti Infect Ther 2009; 7: 12011221.