Methods of hair loss evaluation

Background

Methods currently used in the trichological offices to diagnose hair loss and to measure its severity are the pull test, trichoscopy, the classic trichogram [1], the modified wash test (MWT) [2] and the computer-assisted dermoscopy with a dedicated software (TrichoScan®) [3-5].

Experimental design

Forty-one patients complaining of hair loss were studied, excluding those with scarring alopecia, trichotillomania and alopecia areata.

They were tentatively clinically diagnosed as having androgenetic alopecia (AGA) in the presence of a familiar history and rarefaction in the mid-section of the scalp; as telogen effluvium (TE) in the presence of a history of profuse hair loss with a definite date of onset, no or negligible rarefaction; as AGA + TE when those criteria were overlapping.

Thirty-nine of them were women. Their age ranged between 19 and 82 years (mean 52.6 ± 16.4). Twenty-three (56%) had black hairs (either natural or died), 15 (37%) had brown hairs (either natural or died) and 3 (7%) dark blond hairs (either natural or died). Primary outcomes were the percentage of vellus hairs (VHs) as a diagnostic marker of AGA and the number of shed telogen hairs as a marker of TE.

All were examined with the aid of a videodermoscope equipped with TrichoScan® (Fotofinder, Teachscreen Software, Bad Birnbach, Germany). Within an area close to the vertex (about 1 cm2 wide), the hairs were clipped short and photographed at 20× magnification [5]. AGA was diagnosed in the presence of VH. Two cut-offs were considered: ≥10% and >5%. Although all patients were complaining of hair loss, in the absence of any diagnostic dermoscopic criteria for TE [6], the patients below the mentioned cut-offs were diagnosed as being normal.

In the same day, all subjects studied with TrichoScan® underwent MWT [3]. In brief, after 5 days of abstention from shampooing, they soaped and rinsed their hair in a basin and collected all shed hairs. Hairs were counted and divided into two categories: >3 cm and ≤ 3 cm long. The latter were regarded as being VH [7]. Patients shedding >100 hairs and with <10% VH were diagnosed as having TE, patients shedding ≤100 hairs and with ≥10% VH as AGA, and patients shedding >100 hairs with ≥10% VH as AGA-associated TE. Lastly, patients shedding ≤100 hairs with <10% VH were considered being normal or having TE in remission. For the particular purposes of this study, diagnoses were only three (normal, AGA and TE). Patients with AGA + TE were assigned to either AGA or TE according to the relative severity of the associated disorder.

Data were analysed statistically with Student's t-test and the Cohen κ statistic.

Results

Individual results are shown in Table 1. With MWT, the total count averaged 100.6 ± 107.80, (range 6–431) and the VH percentage was 7.1 ± 6.36. With TrichoScan®, the VH percentage averaged 8.4 ± 4.55. The difference was not significant (t = 1.829; P = .069). None of the patients reached the >20% VH percentage as proposed [5]. Clinical diagnosis and MWT diagnosis (>5% cut-off) were fairly concordant (κ = 0.32). With the ≥10% cut-off, κ was 0.24. Clinical and TrichoScan® diagnosis (>5% cut-off) were fairly concordant, although less satisfactorily (κ = 0.22). With the ≥10% cut-off, κ was fully unsatisfactory (0.09). The concordance between MWT and TrichoScan® diagnoses was unsatisfactory (κ = 0.10).

Conclusions

Modified wash test is a non-invasive procedure, designed for the office, that permits to identify patients with TE, with AGA and with TE + AGA. It provides an estimate of the severity of TE (the total number of hairs actually shed), of AGA (the prevalence of VH) and, of course, of their association, allowing the dermatologist to decide which condition is actually prevalent and which should be treated first. Its reliability, specificity and sensitivity have been found acceptable [8].

Drawbacks are the difficulty in convincing the patients to abstain from shampooing for 5 days, in testing people with curly hairs which entangle during shampooing and in testing young male patients with hairs shorter than 4 cm.

TrichoScan® is a fully automated method for measurement of various parameters including hair density and shaft diameter [3-5].

On the automatic setting, it recognises as VH every hair thinner than 40 μm.

The anagen and telogen hair counts (and their ratio) are supplied only by phototrichogram, which requires two pictures of clipped hairs taken 3 days apart, a procedure usually not accepted by office patients. Besides avoiding the tedious manual identification, the method has the advantage of identifying VH by measuring their diameter, whilst MWT measures their length.

As for drawbacks, TrichoScan® does not appraise the number of shed hairs and, in the absence of specific findings for TE [6], it may misdiagnose TE patients as normal people.

Moreover, VH and early anagen hairs are not distinguished possibly biasing AGA diagnosis. Lastly, because of the camera resolution, the hair thickness detection limit of the software is 16 μm and therefore smaller hairs (as well as fairly coloured hairs) escape measurement [9], unless they are previously died. In fact, in Italy, this procedure is hardly needed, given the natural darker colour of the Italian hairs. Older ladies with white hair tend to dye them, usually in a dark blond or brown colour. In our series, only 3 patients had dark blond hairs, which were easily detectable by TrichoScan®, and none were white haired.

We found that the two methods were weakly in agreement with the clinical diagnosis. However, MWT was better (k = 0.32 vs 0.22) especially, as expected, at detecting TE. Only in 17 patients (41%) were the diagnoses obtained with MWT and TrichoScan® concordant. Moreover, we found that the VH >20% cut-off is excessive, while the 5% cut-off may signal even an early AGA diagnosis.

In conclusion, clinical diagnosis is important, but insufficient to detect AGA correctly. In the office, in the absence of other easy suitable criteria, MWT is a cheap, simple and reliable [8] method readily accepted by patients. Although easy to apply, TrichoScan® is expensive, and in TE and in AGA + TE, it may be even misleading.

References
1
Van Scott E J, Reinertson R P, Steinmuller R. J Invest Dermatol 1957: 29: 197–204.
PubMed,Web of Science® Times Cited: 137

2
Rebora A, Guarrera M, Baldari M et al. Arch Dermatol 2005: 141: 1243–1245.
CrossRef,PubMed,Web of Science® Times Cited: 14

3
Hoffmann R. Eur J Dermatol 2001: 11: 362–368.
CAS

4
Gassmueller J, Rowold E, Frase T et al. Eur J Dermatol 2009: 19: 224–231.
Web of Science® Times Cited: 7

5
Riedel-Baima B, Riedel A. Dermatol Surg 2009: 35: 651–655.
Direct Link:
AbstractFull Article (HTML)PDF(443K)ReferencesWeb of Science® Times Cited: 4

6
Inui S. J Dermatol 2011: 38: 71–75.
Direct Link:
AbstractFull Article (HTML)PDF(577K)ReferencesWeb of Science® Times Cited: 7

7
Rushton D H. Dermatol Clin 1993: 11: 47–53.
CrossRef,PubMed,CAS

8
Guarrera M, Cardo P, Rebora A. G Ital Dermatol Venereol 2011: 146: 289–294.
CAS,Web of Science®

9
López V, Martin J M, Sánchez R et al. J Eur Acad Dermatol Venereol 2011: 25: 1068–1072.
Direct Link:
AbstractFull Article (HTML)PDF(331K)ReferencesWeb of Science®

Source