SEARCH

SEARCH BY CITATION

Keywords:

  • dermatophyte;
  • identification;
  • MALDI-TOF mass spectrometry;
  • onychomycosis

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Questions addressed
  5. Experimental design
  6. Results
  7. Conclusions
  8. Conflict of interest
  9. References
  10. Supporting Information

Abstract:  The incidence and prevalence of onychomycosis are rising worldwide. Common diagnostic techniques often lack sensitivity or specificity. Differentiation between non-infectious nail disorders is frequently not possible. The aim of this study was to establish a better diagnostic routine procedure based on modern mass spectrometric peptide analysis techniques. One hundred and fifty-five nail samples from 145 patients with clinically suspected onychomycosis (n = 96, 62%) and without onychomycosis [e.g. nail psoriasis or nail dystrophy resulting from eczema (n = 59, 38%)] were investigated using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) peptide mass fingerprinting in comparison with standard techniques. We demonstrated that MALDI-TOF MS represents a precise, robust and fast tool in diagnostic investigation of nail disorders, which is superior to common standard methods.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Questions addressed
  5. Experimental design
  6. Results
  7. Conclusions
  8. Conflict of interest
  9. References
  10. Supporting Information

Onychomycosis is a fungal infection with a prevalence between 6.0% and 26.9% in industrialized countries (1,2). The incidence of onychomycosis is increasing worldwide, even in children, in whom onychomycosis was formerly considered as a rare disorder (3,4). It is estimated that both incidence and prevalence will increase in the coming decades, as the number of patients at risk of onychomycosis is rising too (1,2,5). If local or systemic treatment is planned, it is required that onychomycosis is proven by a diagnostic method (6). Furthermore, it is important that nail dystrophy resulting from onychomycosis can be differentiated from that arising from non-infectious nail disorders, e.g. nail psoriasis or nail dystrophy in association with eczema, lichen ruber or traumata (6).

KOH preparation, fungal culture and histological investigation of nail material using periodic acid-Schiff (PAS) staining are standard diagnostic methods. Both KOH and fungal culture often yield false-negative results, especially in pretreated patients. Moreover, results are dependent on the knowledge of the observer. The most remarkable handicap of fungal culture is the long incubation time, which can take up to 3–4 weeks (7).

Questions addressed

  1. Top of page
  2. Abstract
  3. Background
  4. Questions addressed
  5. Experimental design
  6. Results
  7. Conclusions
  8. Conflict of interest
  9. References
  10. Supporting Information

The intention of this study was to compare different standard diagnostic routine procedures (KOH preparation, fungal culture and histological examination including PAS staining) with MALDI-TOF MS in relation to sensitivity, specificity, time-frames and the ability to discriminate between infectious and non-infectious nail disorders.

Experimental design

  1. Top of page
  2. Abstract
  3. Background
  4. Questions addressed
  5. Experimental design
  6. Results
  7. Conclusions
  8. Conflict of interest
  9. References
  10. Supporting Information

Diagnosis was performed using the Species Identification of Animals using MALDI-TOF Mass Spectrometry method (SIAM), a recently developed MALDI-TOF MS-based method (8). The method is exclusively based on the analysis of nail proteins and uses minute amounts of peptides directly derived from tryptic nail digests, without any prior or further clean-up or enrichment steps. Affected samples are identified by comparison of their peptide patterns with known peptide spectra from nail disorders stored in an already existing data base (8). Hereby, correlation distances, a form of Euclidean distances, and deduced parameters are used to measure similarities.

Correlation distance calculations were used for principal coordinates analysis (PCA), a mathematical approach for the two-dimensional visualization of the similarities calculated. The more the spectra are similar among each other the closer they lie together forming clusters. Ideally, onychomycosis samples, samples from healthy probands and those from patients suffering from endogenous diseases are forming three distinct similarity clusters each without common areas. Unknown samples can be allocated by the similarity cluster they are belonging to. The MALDI-TOF MS method and patient characteristics are presented in detail online as Supporting information.

Results

  1. Top of page
  2. Abstract
  3. Background
  4. Questions addressed
  5. Experimental design
  6. Results
  7. Conclusions
  8. Conflict of interest
  9. References
  10. Supporting Information

One hundred and fifty-five nail samples from 145 patients were analysed. Ninety-six (62%) nail samples were investigated under the clinical diagnosis of onychomycosis and 59 (38%) samples under the clinical diagnosis of a non-mycotic nail disorder (e.g. psoriasis, nail eczema or lichen ruber).

KOH preparation was performed in 145 of 155 nail samples. Eighty-nine of 145 (61.4%) of these samples were examined under the clinical diagnosis of onychomycosis. In 30 of these 89 samples (33.7%), KOH preparation was able to identify dermatophytes, yeasts or moulds, i.e. sensitivity = 33.7% (95% CI: 24–44.5).

Mycological culture was performed in 145 of 155 nail samples. Ninety-one of 145 (62.8%) of these samples were cultivated under the clinical diagnosis of onychomycosis. In 26 of these 91 samples (28.6%), mycological culture was able to identify dermatophytes, yeasts or moulds, i.e. sensitivity = 28.6% (95% CI: 19.6–39).

Histology was performed in 131 of 155 nail samples. Eighty-three of 131 (63.4%) of these samples were investigated under the clinical diagnosis of onychomycosis. In 19 of these 83 samples (22.9%), histology was able to identify dermatophytes, yeasts or moulds, i.e. sensitivity = 22.9% (95% CI: 14.4–33.4).

The distances of all test samples (matrix B, see Supporting information) to 26 undoubtedly non-mycotic samples analysed earlier were averaged and form as classifier (or predictor), the basis for the ROC analysis. The gold standard was defined as follows: at least KOH preparation or mycological culture or histology had to be positive. This combination had a sensitivity of 44.8%. Figure 1 shows the result of the ROC analysis of the MALDI-TOF MS method. The area under the curve was 0.61. Using a threshold from 0.266 resulted in a sensitivity of 78.2% (95% CI 65.3–87.7) and a specificity of 37.5% (95% CI: 27.8–48). Comparing with the clinical diagnosis we found a sensitivity of 78.8% (95% CI: 69.4–86.6) and a specificity of 48.3% (95% CI: 35.2–61.6).

image

Figure 1.  ROC curve for MALDI-TOF MS indicating sensitivity and specificity depending on the distance measured between the single nail sample and the undoubtedly non-mycotic reference sample (classifier).

Download figure to PowerPoint

Two-dimensional PCA visualization of the correlation distance data from MALDI-TOF MS investigations resulted in three similarity clusters (Fig. 2). The onychomycosis cluster was clearly defined from the two others. One was the cluster of healthy probands, and the other was the cluster of endogenous diseases. The clusters of healthy probands and endogenous diseases showed an intersection, which indicates flowing transitions between both clusters. Data from non-mycotic diseases can be distinguished from those of healthy probands if they develop disturbances in keratin synthesis or secondary environmental affections like chemical corrosion, while diseases, which are not affecting keratin, cannot be differentiated from healthy probands by the mass spectrometric method presented. In cases, where an endogenous disease was superposed by an onychomycosis infection, the SIAM method allocated the infection due to its enzymatic keratin degradations.

image

Figure 2.  Principal coordinates analysis (PCA) as 2-dimensional visualization of correlation distance data reflecting mass spectral similarities. Correlation distance data are depicted from healthy probands, patients suffering from onychomycoses and from patients suffering from endogenous diseases of nails. Diagnosis data from mass spectrometry analyses were the base for cluster formation. Each cluster is encircled. The more distant two clusters are the more significant is their mass spectral differentiation.

Download figure to PowerPoint

Conclusions

  1. Top of page
  2. Abstract
  3. Background
  4. Questions addressed
  5. Experimental design
  6. Results
  7. Conclusions
  8. Conflict of interest
  9. References
  10. Supporting Information

We showed that MALDI-TOF MS was able to discriminate between onychomycosis and non-fungal nail disorders, whereas KOH preparation and fungal culture are frequently only able to prove or rule out onychomycosis and is in line with preliminary work from our group (8). This is important as only 50% of nail diseases is caused by onychomycosis (9,10). MALDI-TOF MS requires no living or non-living fungal material to prove or to rule out onychomycosis. The novel aspect of this technique is its basis on the detection of solid biochemical characteristics following the activity of mycological infections or non-infectious diseases, which are represented in proteolytic degradation products of native nail proteins. The sensitivity and specificity of different test methods vary markedly between our study and previous studies (6,7,11). One possible explanation for this might be the sampling of nail material by untrained staff or the fact that nearly 50% of our patients was already pretreated with local or systemic antimycotic drugs. As MALDI-TOF MS results are generated by software-assisted analysis and the preparation of nail material is simple, the knowledge of the observer is less important, especially in comparison with other newer methods for the diagnosis of onychomycosis, such as PCR sequencing or PCR-restriction fragment length polymorphism, which need highly skilled staff (12–14). However, test results of these methods are available within 24 h, comparable with MALDI-TOF MS.

Erhard and co-workers were recently been able to identify several dermatophyte species by MALDI-TOF MS (15). The technique described in their manuscript is comparable with our technique, but in contrast to our method they used MALDI-TOF MS for the identification of already grown dermatophytes. This identification is surely more precise concerning species identification, but needs as much time as mycological culture and is more expensive, as costs for mycological culture and MALDI-TOF MS have to be summed up.

In conclusion, we demonstrated that MALDI-TOF peptide mass fingerprint analysis in the diagnostic investigation of nail disorders seems to be a promising new diagnostic technique with a high sensitivity, which is able to discriminate between mycotic and non-mycotic conditions.

References

  1. Top of page
  2. Abstract
  3. Background
  4. Questions addressed
  5. Experimental design
  6. Results
  7. Conclusions
  8. Conflict of interest
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Background
  4. Questions addressed
  5. Experimental design
  6. Results
  7. Conclusions
  8. Conflict of interest
  9. References
  10. Supporting Information

Figure S1. Study flow chart.

Figure S2. Principle of MALDI-TOF analysis.

Figure S3. Comparison of spectra from a healthy person and a patient with onychomycosis.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
EXD_838_sm_FigS1.ppt14KSupporting info item
EXD_838_sm_FigS2.doc180KSupporting info item
EXD_838_sm_FigS3.doc1982KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.