Chapter 5: Principles of Diagnosis

Diagnosis of cutaneous disorders can be done using 2 different methods: (1) visual pattern recognition, a rapid, intuitive, nonanalytic method, or (2) an analytic method using algorithms and decision trees. In actual practice most clinicians use both methods.^1,2

Clinicians utilizing pattern recognition rapidly identify the cutaneous findings and compare them with a set of images stored in their long-term memory (sometimes called the "blink" diagnosis).^2,3 These images are typically from clinical findings seen in previous patients or from pictures in textbooks and other sources. Pattern recognition is more effective in common disorders with typical presentations and in the hands of more experienced clinicians. However, studies have shown that in EKG interpretation⁴ and in dermoscopy,⁵ it can be an effective diagnostic strategy even for less experienced clinicians.

Human beings and all animals are hardwired for visual pattern recognition. Otherwise we could not easily and quickly identify each other, objects, or predators. We know that pattern recognition can be learned, but can it be taught? One of the problems is that much of visual pattern recognition occurs subconsciously.³ An experienced clinician can usually tell a student what cutaneous findings led to his or her diagnosis, but there may be other subtle, but important factors, not easily elucidated, that also contributed to the diagnosis.

The most common features of skin disorders used in pattern recognition include the following:

• The morphology of the primary lesion, its surface changes, color, and size

• Location of lesions

• Configuration of lesions

Many common skin disorders have characteristic features. For example, pink plaques with silvery scale on knees and elbows are characteristic for psoriasis (Figure 5-1). These patterns are covered in more detail in Section Two of this book.

An analytic method for diagnosis is slower and more methodical. It utilizes a step by step evaluation of the patient's history, the physical examination findings, and results of diagnostic tests (eg, potassium hydroxide [KOH] examination and skin biopsies).^2,3 These are used as the basis for searches in differential diagnosis lists or decision tree algorithms. An analytic method is helpful in complex cases with atypical or numerous cutaneous findings and systemic complaints. Section Three of this book contains lists of differential diagnoses of skin diseases in various body regions based on the patient's history, lesion morphology, and laboratory results. It also contains differential diagnosis lists for purpura, pruritus, rash and fever, and leg ulcers.

Both strategies for diagnosis rely heavily on identification and classification of the primary lesion(s). However, there are several pitfalls in the identification of primary lesions. These include:

• Excoriations may alter or partially destroy the primary lesions (Figure 5-2).

• Vesicles, bullae, and pustules may easily break, leaving only erosions or erythema (Figure 5-3). Also, vesicles may develop into pustules as in some case of herpes simplex and zoster.

• Postinflammatory hyperpigmentation and hypopigmentation may obscure the primary lesion (Figure 5-4).

• The examination may take place too early or too late in an evolving skin disease. For instance, many skin rashes present with pink macules in the first 1 to 2 days and then evolve into their more characteristic findings. As an example, herpes zoster could evolve from a pink macule/patch → pink plaque → plaque with vesicles → bullae → crusts → erosions → scars (Figure 5-5).

• Treatment may alter the skin findings, for example, the use of a topical antifungal cream for tinea corporis may remove the characteristic scale or make it difficult to find fungal hyphae on a KOH examination.

• Lastly, many common diseases present with multiple types of primary lesions. For example, atopic dermatitis may present with macules, patches, papules, plaques, vesicles, and pustules and with surface changes that may include scale, crust, lichenification, fissures, erosions, and excoriations.

Identification of the primary lesion(s) helps to categorize rashes. It is important to first consider the broad categories of disease in the differential diagnosis before making a specific diagnosis. The common skin rashes diseases covered in this textbook fall into 4 broad categories.

• Dermatitis

• Papulosquamous diseases

• Urticaria and drug rashes

• Infections (fungal, bacterial, viral)

Most errors in the diagnosis of rashes are made between inflammatory diseases and infectious diseases, for example, misdiagnosing dermatitis as a fungal infection (Table 5-1).⁶ Misdiagnosis is especially problematic in these cases, because the diseases in these 2 categories have different treatments. However, it is important to note that there are some clinical cases in which an inflammatory disease and an infectious process coexist. The following are some examples:

• Atopic dermatitis and staphylococcal skin infection or herpes simplex

• Contact dermatitis and bacterial infection (Figure 5-6)

• Stasis dermatitis and cellulitis

• Dermatitis and molluscum contagiosum

• Dyshidrotic dermatitis and tinea pedis

A study of primary care physicians' errors found that 68% of misdiagnoses could be eliminated by 3 simple diagnostic tests: (1) KOH examination of scale for fungus, (2) Tzanck smear or viral culture for herpes, and (3) skin scraping for scabies.⁶ There are several other laboratory and diagnostic studies such as bacterial cultures and skin biopsies that can be done to confirm diagnoses. However, false-positive and false-negative results can occur due to errors in collection of specimens and in interpretation of the microscopic findings. Therefore, clinical correlation is important. The following are examples of laboratory tests that can be used to confirm diagnoses in various clinical findings:⁷

• KOH examination and/or fungal cultures: Scaling with or without alopecia in preadolescent children, annular scaly plaques at any age, scaling or vesicles on feet, hands, or groin area or body folds, and dystrophy of nails.

• Tzanck smear, polymerase chain reaction (PCR), or viral culture for herpes simplex or varicella zoster virus: Vesicles on an erythematous base, especially on face, genitals, or a unilateral dermatome.

• Scabies scraping: Intensely pruritic papules or vesicles and/or burrows in patients of any age, especially on the hands, feet, extensor extremities, or genitals.

• Bacterial cultures: Cysts or nodules with purulent drainage and crusted plaques surrounded by erythema.

• Blood studies: These are generally not helpful in the diagnosis of common skin disorders, but are helpful in connective tissue disorders. Indirect immunofluorescence done on serum is useful in the diagnosis of pemphigus and pemphigoid.

• Skin biopsies: Histopathologic examination of skin biopsy specimens is useful in the diagnosis of benign and malignant tumors. It is also helpful in the diagnosis of psoriasis, lichen planus, pemphigoid, pemphigus, vasculitis, connective tissue disorders, and many scalp disorders. It is also useful in the diagnosis of fungal infections, but special stains such as the periodic acid Schiff stain (PAS) need to be employed to visualize the fungal elements. Skin biopsies are generally not useful in differentiating between various forms of dermatitis. The diagnostic accuracy of pathology reports is enhanced by the inclusion of detailed history and clinical findings on the biopsy requisition form.^8,9

  Skin biopsies for direct immunofluorescence are diagnostic in pemphigoid, pemphigus, dermatitis herpetiformis, and cutaneous lupus erythematosus.

  There can be differences in interpretation between some pathologists, so as with any laboratory test, clinical correlation is needed.¹⁰ Chapter 7 has more details on skin biopsy procedures.

Many common diseases that are in different broad diagnostic categories may not only mimic each other but also closely resemble skin findings in systemic diseases or diseases with high morbidity (Table 5-2). There are also several skin cancers that are very similar in appearance to benign conditions (Table 5-3).

"Could this be a serious disease?" is a common concern in medicine. Most skin disorders that have high morbidity or are associated with systemic disease have features that should alert a clinician that he or she may be dealing with a serious disorder. Table 5-4 lists some of the clinical findings that could indicate a serious disease.