Onychomycosis: Diagnosis and definition of cure

Richard K. Scher, MD, FACP,a Amir Tavakkol, PhD, Dip Bact,b Bardur Sigurgeirsson, MD, PhD,c

Roderick J. Hay, DM,d Warren S. Joseph, DPM,e Antonella Tosti, MD,f

Philip Fleckman, MD,g Mahmoud Ghannoum, MSc, PhD,h David G. Armstrong, DPM,i

Bryan C. Markinson, DPM,j and Boni E. Elewski, MDk

New York, New York; East Hanover, New Jersey; Reykjavik, Iceland; Belfast, Ireland;

Coatesville, Pennsylvania; Bologna, Italy; Seattle, Washington; Cleveland, Ohio;

North Chicago, Illinois; and Birmingham, Alabama

Until now, there has been no agreement on criteria defining resolution of onychomycosis. Most publishedreports use clinical and mycological cure, which comprises a completely normal-appearing nail plate,and negative nail culture and microscopy results, as the end point for defining success of therapeuticintervention. Reported here is the definition of onychomycosis, which delineates both primary andsecondary criteria for diagnosis of onychomycosis and identifies clinical and laboratory parameters todefine a resolved fungal nail infection. Onychomycosis cure is defined by the absence of clinical signs orthe presence of negative nail culture and/or microscopy results with one or more of the following minorclinical signs: (1) minimal distal subungual hyperkeratosis; and (2) nail-plate thickening. Clinical signsindicative of persistent onychomycosis at the end of the observation period include (1) white/yellow ororange/brown streaksorpatches inor beneath thenail plate; and (2) lateral onycholysiswith subungual debris.Although nail appearance will usually continue to improve after cessation of therapy, the nails may havea persistent abnormal appearance even in cases where treatment has been effective. ( J Am Acad Dermatol2007;56:939-44.)

Onychomycosis is estimated to affect ap-proximately 2% to 13% of the populationof North America and Europe.1-7 Both

physician and patient expectations of treatment

From the Department of Dermatology, Columbia University, New

Yorka; Dermatology Clinical Research, Novartis Pharmaceuticals

Corporation, East Hanoverb; Department of Dermatology, Uni-

versity of Iceland, Reykjavikc; Queens University Belfastd; Ve-

terans Administration Medical Center, Coatesvillee; Department

of Dermatology, University of Bolognaf; Department of Medi-

cine (Dermatology), University of Washington School of Med-

icine, Seattleg; Department of Dermatology, Case Western

Reserve University, Clevelandh; Dr William M. Scholl College

of Podiatric Medicine at Rosalind Franklin University of Medi-

cine and Science, Green Oaks, North Chicagoi; Department of

Orthopedic Surgery, Mount Sinai Medical Center, New Yorkj;

and Department of Dermatology, University of Alabama,

Birmingham.k

Supported by Novartis Pharmaceuticals Corporation.

Disclosure: This report is based on a consensus conference

sponsored by Novartis Pharmaceuticals Corporation. Technical

assistance from CPE Communications is appreciated. All con-

tributors have received honoraria from Novartis Pharmaceuti-

cals Corporation for their participation in the consensus

conference meeting that formed the basis for this article.

Dr Tavakkol is Director of Dermatology Clinical Research at

Novartis Pharmaceuticals Corporation. All authors received

honoraria from Novartis and are consultants/advisors, speakers,

and investigators for Novartis. Dr Scher is a consultant, and

investigator, and received honoraria and grants from Barrier.

He is also an advisory board member, consultant, and received

honoraria from Stiefel. Dr Sigurgeirsson is an investigator,

consultant, and speaker for Galderma and has received grants,

outcome are influenced by the widely varying per-ceptions of what constitutes cure of onychomycosis.Yet, physicians often do not discuss with theirpatients the likely end result of therapy.

honoraria, and salary from this company. He is also an advisory

board member, investigator, and consultant for Stiefel and has

received grants and honoraria from this company. Dr Hay is

an advisory board member receiving honoraria from Barrier.

Dr Tosti is an advisory board member receiving honoraria

from both Stiefel and Galderma. Dr Ghannoum is an investigator

and speaker receiving grants and honoraria from Pfizer;

and investigator and speaker receiving grants and honoraria

from Enzon; an investigator and consultant receiving grants

and honoraria from Schering-Plough; and investigator and

speaker receiving grants and honoraria from Merck; an inves-

tigator receiving grants and honoraria from Vicuron; and a

consultant receiving honoraria from NexMed. Dr Armstrong is

an advisory board member and investigator for Lilly. He is also

an advisory board member of, investigator for, and receives

honoraria from Pfizer. Dr Markinson is a speaker, consultant,

and advisory board member receiving honoraria and stock

options for Bradley Pharmaceuticals and is an advisory board

member receiving honoraria from Stiefel. Dr Elewski is an

advisory board member receiving honoraria from Anacor and

Stiefel and an investigator for Barrier and Novartis.

Accepted for publication December 21, 2006.

Reprint requests: Boni E. Elewski, MD, Department of Dermatology,

University of Alabama, 700 18 St S, Suite 414, Birmingham,

AL 35233. E-mail: beelewski@aol.com.

Published online February 20, 2007.

0190-9622/$32.00

ª 2007 by the American Academy of Dermatology, Inc.

doi:10.1016/j.jaad.2006.12.019

939

J AM ACAD DERMATOL

JUNE 2007

940 Scher et al

Further, clinicians base their therapeutic decisionson results of clinical trials. However, the currentclinical trial data are often difficult to interpret andcompare because there have been numerous defini-tions of ‘‘cure’’ and different measures used to assesstreatment efficacy. Most published studies use dif-ferent combinations and definitions of mycologicalcure (often defined as negative potassium hydroxide[KOH] microscopy and culture results), clinical cure(often defined as a percentage [eg, 80%-100%] of nailplate that is visibly clear of infection), and completecure (mycological plus clinical cure) to define efficacyand cure. Large-scale and adequately powered clini-cal trials using the most stringent efficacy criteria oftenshow low rates of therapeutic response and com-plete cure than seen in clinical practice or publishedstudies. The lack of consistency in defining and meas-uring cure is one reason for the disparity betweenthe results of therapy, as recorded in drug trials, andexperience from clinical practice. For example, inan analysis of 26 published clinical studies for oraltreatment of toenail onychomycosis, a complete curewas achieved in only 25% to 50% of patients receivingstandard courses of therapy,8 and complete cure ratesin the range of 14% to 38% have been reportedin the prescribing information of Food and DrugAdministrationeapproved oral agents.9,10 In clinicaltrials, efficacy assessments are often based on finalevaluations at 48 to 52 weeks, but a toenail may notgrow fully for up to 78 weeks.11,12 Although onewould not expect all studies to be the same, variationsin study designs might also account for some dispar-ities in cure rate.

In addition, many clinical studies base treatmentsuccess on the progress of a single target toenail,whereas in clinical practice, patients and physiciansevaluate the potentially different responses of alltoenails. Although mycological cure in a target toe-nail usually corresponds with similar trends in the

Table I. Diagnosis of onychomycosis causedby dermatophytes

ClinicalPrimary criteria

White/yellow or orange/brown patches or streaksSecondary criteria*

OnycholysisSubungual hyperkeratosis/debrisNail-plate thickening

LaboratoryPositive microscopic evidencePositive culture of dermatophyte

*Tinea pedis often occurs concomitantly with pedal onychomycosis,

and tinea manuum with infected fingernails.

other nails, in some cases one or more of the othertoenails will continue to show signs of infection.13

DIAGNOSIS OF ONYCHOMYCOSISThe criteria for the diagnosis of dermatophyte

onychomycosis, including both laboratory and clin-ical features, are summarized in Table I.

Dependence on culture of an organism alone isnot sufficient for the diagnosis of infectionea fungusisolated from a normal nail does not demonstrateinfection. The reverse is also trueean abnormal nailwithout mycological confirmation is insufficient tomake an accurate diagnosis of onychomycosis. Theaccurate diagnosis can be made only when bothpositive laboratory and clinical criteria are present.

Tinea pedis and tinea manuum offer clinical cluesof infection because they often occur concomitantlywith pedal onychomycosis. Certain nail changes maybe nonspecific. Onycholysis, for example, may resultfrom trauma but could also be seen in psoriasis, as issubungual hyperkeratosis. Nail-plate thickening isalso relatively nonspecific because it may be associ-ated with trauma, onychogryphosis, lichen planus,and psoriasis. Other nail anomalies that are generallyunrelated to onychomycosis include longitudinal ortransverse ridging, pits, onychoschizia, and drynessof the nail plate. Surface leukonychia may be seen insome forms of onychomycosis, including white su-perficial onychomycosis and proximal white subun-gual onychomycosis, but is otherwise nonspecific.These factors make diagnosis of onychomycosis onclinical grounds alone difficult and correlation withmycological evidence critical. It should be noted that avariety of clinical signs present at the initial clinicalevaluation indicate a poor overall prognosis for ulti-mate cure. These factors are listed in Table II.14,15

Definitive laboratory criteria include positivemicroscopic evidence of septate hyphae and/orarthroconidia (KOH preparation, Calcofluor white,Sigma-Aldrich, St Louis, Mo), periodic acideSchiff,and/or biopsy, and positive fungal culture findingsfor dermatophytes (Trichophyton, Epidermophyton,

Table II. Poor prognostic factors

1. Areas of nail involvement [ 50%2. Significant lateral disease3. Subungual hyperkeratosis [ 2 mm4. White/yellow or orange/brown streaks in the nail

(includes dermatophytoma14)5. Total dystrophic onychomycosis (with matrix

involvement)6. Nonresponsive organisms (eg, Scytalidium mold)7. Patients with immunosuppression8. Diminished peripheral circulation

J AM ACAD DERMATOL

VOLUME 56, NUMBER 6

Scher et al 941

Table III. Proposed definitions of cure when assessing patients with onychomycosis in clinical trials

Criteria for cure:A) 100% Absence of clinical signs of onychomycosis (mycology not required)

ORB) Negative mycological laboratory results with one or more of following clinical signs

i) Distal subungual hyperkeratosis or onycholysis leaving less than 10% of nail plate affectedii) Nail-plate thickening that does not improve with treatment because of comorbid condition

Criteria for noncure:A) Presence of positive mycological results

ORB) Any one of the 4 clinical signs, even in the presence of negative mycological results

i) Residual major changes ([10%) of the nail plate compatible with dermatophyte infectionii) White/yellow or orange brown/patches or streaks in or beneath the nailiii) Lateral onycholysis with debris in an otherwise clear nail plateiv) Hyperkeratoses on the lateral nail plate/nailfold edge (Fig 1, D)

or Microsporum species) or certain nondermato-phyte nail pathogens (eg, Scytalidium dimidiatumand S hyalinum). Candida albicans can occasionallybe a pathogen in fingernail disease, but clinicalcorrelation is critical. Other nondermatophyte nailpathogens (certain species of Acremonium,Alternaria, Aspergillus, Fusarium, Onychocola, andScopulariopsis) must be isolated from sequentialspecimens to prove origin, and correlation withdirect microscopy and clinical changes is required.Mycology test results should be used to confirm theclinical diagnosis and it should be noted that absenceof a proof is not a proof of absence.

Successful eradication of the fungus may leavethe nail abnormal and the residual changes may betotally unrelated to infection (eg, onychoschizia) or aresult of damage to the nail unit from long-standingdisease (eg, onycholysis).

If onychomycosis is suggested based on clinicalobservation, diagnostic laboratory tests should beperformed. If these produce negative findings, theyshould be repeated. Clinical manifestations of othernail disordersesuch as psoriasis, neosplasms, andlichen planusemay mimic those of onychomycosisbut can be diagnosed by nail-unit biopsy.16-18

PRACTICAL CONSIDERATIONS FORASSESSING CURE

Although the goal of patients seeking treatmentis almost always a normal-appearing nail, it has beensuggested that cure of all 10 toenails is a desiredclinical outcome, but the latter may be unattainable.It is important that patients understand the distinc-tion between cure of their nail infection and grossappearance of the affected (now treated) nail, be-cause some residual change is likely after chronicinfection. Assessment of cure should not rely entirelyon visual appearance, and patients and physicians

need to have realistic expectations of successfultreatment outcome.

Mycological cure signifies that the fungal infectionhas been successfully treated and has resolved, but itwill not necessarily result in a 100% normal nail.19

Both intrinsic and extrinsic factors may influence theappearance of the nail. Trauma to the nail unit,particularly the bed and plate, may precede thedevelopment of onychomycosis and even a return tomycological negativity will not necessarily producea normal-appearing nail. In severe cases of onycho-mycosis, up to 10% of the nail surface is likely toremain abnormal in appearance even when mycol-ogy indicates a cure of fungal infection.20

CLINICAL SIGNS ACCEPTABLE IN CUREDCASES AT THE END OF THEOBSERVATION PERIOD

Several minor clinical signs that, when combinedwith negative mycology laboratory results, may bepresent in a patient who has been successfully curedare summarized in Table III. Although repeatedmycological testing is not necessary, if clinicalsuggestion of infection is strong, a second sampleshould be taken to confirm the negative mycologicalresults. The presence of minor clinical signs (TableIII) and positive mycological laboratory results indi-cates that the nail is not cured, and further treatmentand/or appropriate follow-up may be indicated. Incontrast, in the absence of clinical signs of onycho-mycosis, mycology laboratory assessments (eg, ei-ther KOH or cultures) are not required to validatecure, although some physicians consider mycologi-cal confirmation optimal practice.

Certain clinical signs may require diagnostic my-cology laboratory tests to confirm cure. Distal ony-cholysis that has improved from baseline but affectsmore than 10% of the nail might be a clinical concern.

J AM ACAD DERMATOL

JUNE 2007

942 Scher et al

Fig 1. Examples of pretreatment and posttreatment of onychomycosis. A, Patient (36-year-oldman) who recently experienced subtle change of big toenail. There is mild distal onycholysiswith bilateral edge involvement. In this case, it was impossible to make diagnosis ofonychomycosis without laboratory confirmation. Microscopy produced positive results andTrichophyton rubrum was cultured from lateral edge. B, Patient (34-year-old man) withonychomycosis of distal subungual type with approximately 30% distal involvement. There isonycholysis with mild distal hyperkeratosis and nail-plate fragility. Microscopy revealedpositive findings and T rubrum was cultured. C, Patient (29-year-old man) with distal lateralsubungual onychomycosis (DLSO). Nail is 90% involved. Nail plate is thickened and subungalhyperkeratosis is apparent. This nail has several negative prognostic factors and is likely to bedifficult to treat. On first sample, microscopy revealed positive findings, but culture producednegative results. However, T rubrum was grown on culture when second, more proximalsample was obtained. D, Patient (46-year-old woman) treated with standard course of oralantifungal. At baseline, nail had 50% area involvement and matrix was affected. Photographwas taken 18 months after treatment was initiated. Patient was satisfied and, on examination,only slight distal onycholysis and scaling/hyperkeratosis of lateral nailfolds was noted. Whenedge was trimmed, very little subungual debris could be found. Material was sent for culture;the microscopy revealed positive findings and T rubrum was cultured. E, Patient (64-year-oldman) with severe onychomycosis before treatment was initiated. Nail displays several negativeprognostic signs. Whole nail plate is involved; nail is thick; dermatophytoma is seen at leftlateral edge; and there is severe hyperkeratosis at lateral edge. F, Patient was treated withstandard course of oral antifungal. T rubrum could be cultured from nail up to 3 months aftertreatment was initiated and microscopy revealed positive findings for up to 1 year. Both cultureand microscopy results remained negative at 18 and 24 months. Clinically, there is slightdiscoloration at distal edge of nail and minimal onycholysis is noted. These signs are minor andare compatible with cure because both microscopy and culture results were negative. G, Beforetreatment was initiated, 100% of nail plate was involved. Patient (37-year-old woman) wastreated with standard course of oral therapy. T rubrum was cultured and microscopy revealedpositive findings up to 6 months after treatment was initiated. Both parameters remainednegative thereafter at 9, 12, 18, and 24 months. Clinical status improved continuously duringand after treatment. Photograph shows nail at 12 months. Distal onycholysis can be seen, andconsiderable hyperkeratosis was noted when distal nail plate was removed. These signs arecompatible with cure only in light of negative microscopy and culture results. Hyperkeratosiswas not apparent at 18 and 24 months, but mild distal onycholysis could still be seen.

J AM ACAD DERMATOL

VOLUME 56, NUMBER 6

Scher et al 943

Lateral nail plate/nailfold edge hyperkeratosis mayalso warrant additional tests.

RECURRENCE OF ONYCHOMYCOSISRecurrence (relapse or reinfection) of onychomy-

cosis is not uncommon, with reported rates rangingfrom 10% to 53%.21-23 Recurrence often implies that,although the clinical signs have resolved, eithermycological cure was not achieved with the initialtreatment or a new infection has developed during orimmediately after the treatment period. Thus, ensur-ing that mycological cure is achieved is important inquestionable cases, so that relapse may be avoided.Positive mycological laboratory findings (KOH, pe-riodic acideSchiff, or culture) are not consistent withcure. Examples of treated onychomycosis resultingin cure with and without residual clinical signs, anduncured onychomycosis, are depicted in Fig 1.

TIMING OF OBSERVATIONSWhen evaluating newly formed nails for clearance

of infection, it is important to remember that thetime required for visible clearance of the infection isdependent on the rate of nail growth. Generally, ittakes 12 to 18 months for a newly formed toenailplate, and 4 to 6 months for a fingernail to replace adiseased nail.24,25 Several factors may influence therate of nail growth, including the age of the patient(the rate of linear nail growth decreases by 50%during a normal life span26), concomitant condi-tions,24 and certain medications that may causeeither an increase or decrease in nail growth.24,27

Immediately after completion of therapy with oneof the newer, long-lasting, systemic antimycotics, thenail usually does not appear to be clear of infection. Aproximal area of normal-appearing nail, representingnewly formed nail plate that is devoid of infection,may be present. Visible clearance of the infection willoccur after the process of nail-plate turnover iscomplete.25 Thus, the concept of cure should allowfor progress and growth from baseline. The same curecriteria shouldbe appliedduring follow-upevaluations.

CONCLUSIONSA nail plate with a normal appearance is not

always attainable after a successful therapeutic ony-chomycosis regimen. Permanent nail anomalies maypersist after elimination of the fungal pathogen inspite of the resolution of infection. Clinical andmycological criteria are important to ascertain boththe diagnosis and resolution of onychomycosis.

REFERENCES

1. Elewski BE, Charif MA. Prevalence of onychomycosis in pa-

tients attending a dermatology clinic in northeastern Ohio for

other conditions. Arch Dermatol 1997;133:1172-3.

2. Roberts DT. Prevalence of dermatophyte onychomycosis in

the United Kingdom: results of an omnibus survey. Br J Dermatol

1992;126(Suppl):23-7.

3. Sais G, Jucgla A, Peyri J. Prevalence of dermatophyte onycho-

mycosis in Spain: a cross-sectional study. Br J Dermatol

1995;132:758-61.

4. Heikkila H, Stubb S. The prevalence of onychomycosis in

Finland. Br J Dermatol 1995;133:699-703.

5. Svejgaard EL, Nilsson J. Onychomycosis in Denmark: preva-

lence of fungal nail infection in general practice. Mycoses

2004;47:131-5.

6. Sigurgeirsson B, Steingrimsson O, Sveinsdottir S. Prevalence

of onychomycosis in Iceland: a population-based study. Acta

Derm Venereol 2002;82:467-9.

7. Gupta AK, Jain HC, Lynde CW, Watteel GN, Summerbell RC.

Prevalence and epidemiology of unsuspected onychomycosis

in patients visiting dermatologists’; offices in Ontario, Canadae

a multicenter survey of 2001 patients. Int J Dermatol 1997;36:

783-7.

8. Epstein E. How often does oral treatment of toenail onycho-

mycosis produce a disease-free nail? An analysis of published

data. Arch Dermatol 1998;134:1551-4.

9. Sporanox (itraconazole) [prescribing information]. Titusville,

NJ: Janssen; 2004.10. Lamisil (terbinafine HCl) [prescribing information]. East Hano-

ver, NJ: Novartis; 2004.

11. Werschler WP, Bondar G, Armstrong D. Assessing treatment

outcomes in toenail onychomycosis clinical trials. Am J Clin

Dermatol 2004;5:145-52.

12. De Cuyper C. Long-term outcomes in the treatment of toenail

onychomycosis. Br J Dermatol 1999;141:15-20.

13. Gupta AK, Ryder J, Summerbell RC. Comparison of efficacy

criteria across onychomycosis trials: need for standardization.

Int J Dermatol 2003;42:312-5.

14. Roberts DT, Evans EG. Subungual dermatophytoma compli-

cating dermatophyte onychomycosis. Br J Dermatol 1998;138:

189-90.

15. Scher RK, Baran R. Onychomycosis in clinical practice factors

contributing to recurrence. Br J Dermatol 2003;149(Suppl):5-9.

16. Weinberg JM, Koestenblatt EK, Tutrone WD, Tishler HR,

Najarian L. Comparison of diagnostic methods in the eval-

uation of onychomycosis. J Am Acad Dermatol 2003;49:

193-7.

17. Elewski BE, Hay RJ. Update on the management of onycho-

mycosis: highlights of the third annual international summit

on cutaneous antifungal therapy. Clin Infect Dis 1996;23:

305-13.

18. Trepanier EF, Amsden GW. Current issues in onychomycosis.

Ann Pharmacother 1998;32:204-14.

19. Elewski BE. A full ‘‘cure’’ for onychomycosis is not always

possible. Arch Dermatol 1999;135:852-3.

20. Hay RJ. The future of onychomycosis therapy may involve

a combination of approaches. Br J Dermatol 2001;145(Suppl):

3-8.

21. Sigurgeirsson B, Olafsson JH, Steinsson JB, Paul C, Billstein S,

Evans EG. Long-term effectiveness of treatment with terbi-

nafine vs itraconazole in onychomycosis: a 5-year blinded

prospective follow-up study. Arch Dermatol 2002;138:

353-7.

22. De Doncker PD, Gupta AK, Marynissen G, Stoffels P, Here-

mans A. Itraconazole pulse therapy for onychomycosis and

dermatomycoses: an overview. J Am Acad Dermatol 1997;37:

969-74.

23. Villars VV, Jones TC. Special features of the clinical use of oral

terbinafine in the treatment of fungal diseases. Br J Dermatol

1992;126(Suppl):61-9.

J AM ACAD DERMATOL

JUNE 2007

944 Scher et al

24. Geyer AS, Onumah N, Uyttendaele H, Scher RK. Modulation

of linear nail growth to treat diseases of the nail. J Am Acad

Dermatol 2004;50:229-34.

25. Del Rosso JQ. Advances in the treatment of superficial fungal

infections: focus on onychomycosis and dry tinea pedis. J Am

Osteopath Assoc 1997;97:339-46.

26. Orentreich N, Markofsky J, Vogelman JH. The effect of aging

on the rate of linear nail growth. J Invest Dermatol 1979;

73:126-30.

27. Jones HE, Zaias N. Double-blind, randomized comparison of

itraconazole capsules and placebo in onychomycosis of toe-

nail. Int J Dermatol 1996;35:589-90.