Clinical dermatology Original article

doi: 10.1111/j.1365-2230.2006.02305.x

Interobserver reliability of the Nail Psoriasis Severity Index

_
zkan
1,2 S . Aktan, T. Ilknur,
C. Akin and S . O
_
Department Of Dermatology, Dokuz Eylul University School of Medicine, Izmir,
Turkey

Summary

Background. Because the Psoriasis Area and Severity Index (PASI) does not consider
the severity of nail disease, a scale that assesses the extent of involvement of psoriatic
nails is needed. A new grading system, the Nail Psoriasis Severity Index (NAPSI) has
been proposed.
Aims. The purpose of this study was to assess the interobserver reliability of NAPSI.
Methods. The nail features of 25 patients with psoriasis with nail involvement
were evaluated and graded by three dermatologists for total NAPSI scores and nail
scores. The quadrants of all nails were examined for the presence of matrix and
bed features. Total NAPSI score (0160) of patients and nail score (032) of the
individual nails were calculated. Interobserver reliability assessments were performed
by computing intraclass correlation coefficients (ICC; two-way mixed model, consistency definition).
Results. The ICC(3,1) results for total NAPSI score and nail score were found to be
0.781 and 0.649, respectively. The ICC(3,1) for nail-bed and nail-matrix features
were 0.869 and 0.584, respectively, in the total NAPSI scoring system, and 0.705 and
0.603, respectively, in the nail scoring system.
Conclusion. Moderate to good agreement of scoring with the NAPSI was determined
among the observers in this study. Our results suggest that scoring for nail-bed features
seems to be more reliable than scoring for nail-matrix features.

Introduction
Psoriasis is a chronic inflammatory disease that affects
skin and nails. Nail changes can be observed in up to
55% of patients with psoriasis, and nail psoriasis is
estimated to affect 8090% of patients with psoriasis at
some point in their lives.13 There is a strong association
between the duration of skin lesions and nail psoriasis,
and nail involvement has been reported to restrict
patients activities.4 The clinical features of nail psoriasis
are related to the portion of the nail unit affected by the
disease, with the main clinical features being pitting,
onycholysis, discolouration, oily spots, nail thickening,

Correspondence: Dr Sebnem Aktan, Dermatoloji Anabilim Dali, Dokuz Eylul

Universitesi Tip Fakultesi, Inciralti, Izmir, Turkey.
E-mail: sebnem.aktan@deu.edu.tr
Conflict of interest: none declared.
Accepted for publication 13 September 2006

subungual hyperkeratosis, psoriatic paronychia, transverse ridging and Beaus lines, nail loss onychomadesis,
and cessation of nail production.3 The Psoriasis Area
and Severity Index (PASI) scale rates the severity of skin
psoriasis, but it does not consider the severity of nail
disease, thus a scale for the assessment of nail psoriasis
is needed. Recently, a new grading system, the Nail
Psoriasis Severity Index (NAPSI), which assesses the
extent of the involvement of the psoriatic nail unit, has
been proposed by Rich and Scher.5 Because this scale
would be useful in clinical trials evaluating different
treatment modalities for psoriatic nails, we aimed to
investigate the interobserver reliability of NAPSI.

Materials and methods

Consecutive patients with psoriasis attending our dermatology outpatient clinic, who also had nail features, were
included in the study. In cases where there was suspicion
of a fungal infection, nail clippings were sent for direct

 2006 The Author(s)

Journal compilation  2006 Blackwell Publishing Ltd Clinical and Experimental Dermatology, 32, 141144

141

Interobserver reliability of the Nail Psoriasis Severity Index S . Aktan et al.

microscopy and culture, and patients with onychomycosis were excluded. In addition, psoriatic arthritis and
pustular psoriasis of the nails were not included in this
study. None of the patients were having systemic antipsoriatic treatment or using a specific topical therapy for
their psoriatic nails at the time of evaluation.
Three dermatologists [SA (observer 1), CA (observer
2), and TI (observer 3)] reviewed the Rich and Scher
paper on the NAPSI scoring system, and used the
standard NAPSI sheet as suggested by the authors. Each
patient was evaluated by the observers on the same day,
under the same conditions, in a well-illuminated room,
and under direct vision. All fingernails and toenails of
the patients were scored, and each observer was blinded
to the scoring of the other observers.
The study population comprised 25 patients who
fulfilled the inclusion criteria (16 men and 9 women,
with a mean SD age of 50.8 11.3; range 2875),
Mean SD PASI score was 15.4 9.1 (range 3.2
34.2) and duration of psoriasis in the group was
18.9 9.4 years (range 247). Chronic plaque-type
psoriasis was the most common clinical form, affecting
21 patients (84%). Three patients (12%) presented with
erythrodermic psoriasis while one patient (4%) had
palmoplantar psoriasis. All nails of all 25 patients were
evaluated. To grade the nails according to the NAPSI,
the nail was divided into quadrants by imaginary
horizontal and longitudinal lines. Each quadrant of
the nail was evaluated for nail-matrix disease (pitting,
leukonychia, red spots in the lunula, crumbling) and for
nail-bed disease (onycholysis, splinter haemorrhages,
subungual hyperkeratosis, salmon-patch dyschromia),
and scored 0 for the absence and 1 for the presence of
any of these nail-matrix or nail-bed findings. The total
nail score (matrix + bed) for all quadrants of the 20 nails
of each patient (0160) was referred to as the patients
total NAPSI score, and was calculated for each of the 25
patients enrolled in the study.
Considering the nail, but not the patient, as the unit of
evaluation for NAPSI score, Rich and Scher5 have also
proposed rating each quadrant with a score of 1 for the
presence of each of the nail-matrix and nail-bed findings
described above. Using this method, the total score of
each nail was computed by summing the (matrix + bed)
scores of all the quadrants of that nail (032) and
referred to as the nail score. Nail score was calculated for
each nail in our study group (n 500 nails).

(two-way mixed-effect model consistency definition),

using SPSS software (version 13.0). The total NAPSI
score (0160) for each patient (n 25), the nail scores
for all nails (032; n 500 nails), and the nail
scores (032) for hands and for feet separately, were
computed.

Results
The total of the points rated by all three observers for
each feature in all nails (n 500), and their percentages are shown in Fig. 1. Means of nail-matrix, nail-bed
and total NAPSI scores are shown in Fig. 2. ICC of total
NAPSI score were computed to assess the interobserver
reliability. ICC(3,1) was 0.781, which indicated a moderate to good agreement (Table 1). The means of the
nail-matrix, nail-bed and nail scores of all nails (n
500) are shown in Fig. 3. ICC(3,1) for nail scores in this
group was found to be 0.649, which was also moderate
to good agreement among the observers (Table 2).

Oil drop
901; 21%
Pitting
1355; 31%

Leukonychia
11; 0,3%
Red spots in lunula
16; 0,4%
Crumbling
187; 4%
Onycholysis
395; 9%
Splinter
haemorrages 37; 1%

Subungual
hyperkeratosis
1409; 34%

Figure 1 Total number and percentage of the points rated by the

three observers for each feature in all nails (n 500).

60
51.1
46.2 46.5

50
40
30
20

20.4

23.4

25.8

Rater 1
Rater 2
Rater 3

29.3 27.7

17.2

10
0

Statistical analysis

Nail matrix score

Interobserver reliability assessments were performed

by computing intraclass correlation coefficients (ICC)

142

Nail bed score

Total NAPSI score

Figure 2 Means of nail-matrix, nail-bed and total NAPSI scores

(n 25 patients).

 2006 The Author(s)

Journal compilation  2006 Blackwell Publishing Ltd Clinical and Experimental Dermatology, 32, 141144

Interobserver reliability of the Nail Psoriasis Severity Index S . Aktan et al.

Table 1 Intraclass correlation coefficients (ICC) for nail-matrix,

nail-bed and total NAPSI scores (n 25 patients).

Score

ICC(3,1)

95% CI

Nail matrix
Nail bed
Total NAPSI

0.584
0.869
0.781

0.3590.769
0.7650.935
0.6250.888

NAPSI, Nail Psoriasis Severity Index.

3
2.32

2.5

2.56

2.31

2
Rater 1
Rater 2
Rater 3

1.46 1.40

1.5
1.02 0.86

1.16

1.29

1
0.5
0
Nail matrix score

Nail bed score

Nail score

Figure 3 Means of nail-matrix, nail-bed, and nail scores of all nails

(n 500 nails).

Interobserver reliability for nail scores was assessed

for fingernails and toenails separately. ICC(3,1) for nail
scores of fingernails (n 250) and toenails (n 250)
were computed as 0.659 and 0.637, respectively, both
indicating moderate to good agreement among the
observers (Table 2).

Discussion
A nail psoriasis-specific scoring system is necessary to
standardize assessment of nail psoriasis to assist reliable
numeric measurement of response to different treatment
modalities and to observe patients longitudinally in
clinical trials. Recently, several grading systems for nail
psoriasis have been proposed. The NAPSI was proposed
as a grading system for nail psoriasis to be used either to
rate all of the nails of a patient or to rate the target nail,
which were referred to as the total NAPSI score and the

nail score, respectively. Baran6 proposed another nail

psoriasis severity index in which pitting, Beaus lines,
subungual hyperkeratosis and onycholysis are scored
separately. He suggested evaluating both the scoring of
each sign and a global score calculated by summing the
severity of each sign for each digit. There are noticeable
differences between the NAPSI and Barans nail severity
scale. According to Barans index, the number of pits,
transverse grooves and the measure of the nail thickness with a calliper are determined, and onycholysis is
evaluated by the assessment of the nail in eight
portions. Splinter haemorrhages, which are often traumatic, are not taken into consideration in Barans
severity index. In contrast, splinter haemorrhage is
taken into account in the NAPSI as one of the features
of nail-bed psoriasis, but Beaus lines are not scored. In
our study, the most common nail feature observed was
subungual hyperkeratosis. Pitting was the second most
frequent abnormality, followed by salmon patches,
onycholysis and crumbling. Splinter haemorrhages,
red spots in the lunula and leukonychia were identified
infrequently (Fig. 1). Thus, in view of the incidence of
each of the nail features in this study, revising the
scoring system by excluding some of these uncommon
features would not seem to have any significant effect
on clinical practice.
Recently, a modified target NAPSI for target-nail
assessment was proposed by Parrish et al.7 They
suggested giving each parameter a degree of gradation
from 0 to 3 (0 none, 1 slight, 2 moderate, 3
severe), as for the PASI, in order to obtain a more
sensitive system for assessing nail changes in response
to therapy.
The interobserver reliability for total NAPSI score and
nail scores was investigated in this study; the interobserver reliability for nail-bed scores seemed to be better
than that for nail-matrix scores, as shown by the ICC for
the nail-matrix and nail-bed scores of the toe nails in
particular. This difference may be due to more difficulties met in evaluating the features of NAPSI in the
quadrants of the relatively small toenail surfaces. In
addition, because matrix features such as longitudinal

Table 2 Intraclass correlation coefficients (ICC) for nail-matrix, nail-bed and nail scores.
All nails (n 500)

Fingernails (n 250)

Toenails (n 250)

Score

ICC(3,1)

95% CI

ICC(3,1)

95% CI

ICC(3,1)

95% CI

Nail matrix
Nail bed
Nail

0.603
0.705
0.649

0.5580.646
0.6670.739
0.6070.688

0.552
0.686
0.659

0.4830.618
0.6300.737
0.6010.714

0.303
0.690
0.637

0.2240.384
0.6350.741
0.5750.694

 2006 The Author(s)

Journal compilation  2006 Blackwell Publishing Ltd Clinical and Experimental Dermatology, 32, 141144

143

Interobserver reliability of the Nail Psoriasis Severity Index S . Aktan et al.

ridges and transverse grooves are not included in the

NAPSI, possible variations in the categorization of these
features might have caused a difference in the scoring of
nail-matrix disease.
Although we did not observe our patients longitudinally, we believe that evaluating the patients using only
the total NAPSI score would have the limitation of
missing de novo lesions on initially healthy nails during
the surveillance of patients after the commencement of
treatment periods, especially when the patients are
followed up by different observers. This limitation is
particularly true for topical treatment modalities in
which the treatment is applied only to the nails with
psoriasis. In our opinion, target nail scores should be
used along with the total NAPSI score in the evaluation
of a patient with nail psoriasis, thus we evaluated the
interobserver reliability of both the total NAPSI scores
and the nail scores in our study group.
In summary, moderate to good agreement of scoring
with the NAPSI was found among the observers in this
study. Further studies comparing and re-evaluating the
different proposed nail severity scales are needed.

144

References
1 de Berker DAR, Baran R et al. The nail in dermatological
diseases. In: Baran and Dawbers Diseases of the Nails and
Their Management (Baran R, Dawber RPR, de Berker DAR
et al, eds), 3rd edn. Oxford: Blackwell Science Ltd, 2001:
17293.
2 de Berker D. Management of nail psoriasis. Clin Exp
Dermatol 2000; 25: 35762.
3 de Berker D. Diagnosis and management of nail psoriasis.
Dermatol Ther 2002; 15: 16572.
4 de Jong EM, Seegers BA, Gulinck MK et al. Psoriasis of
the nails associated with disability in a large number of
patients: results of a recent interview with 1728 patients.
Dermatology 1996; 193: 3003.
5 Rich P, Scher RK. Nail psoriasis severity index: a useful tool
for evaluation of nail psoriasis. J Am Acad Dermatol 2003;
49: 20612.
6 Baran RL. A nail psoriasis severity index. Br J Dermatol
2004; 150: 5689.
7 Parrish CA, Sobera JO, Elewski BE. Modification of the
nail psoriasis severity index. J Am Acad Dermatol 2005; 53:
7456.

 2006 The Author(s)

Journal compilation  2006 Blackwell Publishing Ltd Clinical and Experimental Dermatology, 32, 141144