The normal urothelium, as mentioned previously, is composed of three cell types: basal, intermediate, and superficial cells arranged in three to seven cell layers, with the thickness varying depending on the state of distention (Fig. 2.1) (efigs 2.1-2.7). The nuclei of basal cells are small and hyperchromatic, and the cells usually form a single layer. The intermediate cells constitute the bulk of the urothelium, the cells are oriented perpendicular to the basement membrane, and the nuclei are usually oval to round in tissue sections. Nuclear grooves are frequently identified. The cytoplasm is clear to amphophilic. The superficial or umbrella cells are arranged parallel to the basement membrane and are large, with one cell typically spanning or covering several intermediate cells (like an umbrella). The nucleus is small, and cytoplasm is voluminous and clear or eosinophilic. Mild degrees
of architectural variation (apparent loss of polarity) but without cytologic atypia are considered acceptable to be designated as normal (Figs. 2.2, 2.3). This is often due to tangential or thick sectioning.

Nonkeratinizing squamous epithelium metaplasia, particularly in the area of the trigone, is a common finding in premenopausal women and is responsive to estrogen production (efigs 2.8-2.12). This type of squamous mucosa is characterized by abundant intracytoplasmic glycogen and lack of keratinization, making it histologically similar to vaginal or cervical squamous epithelium (Fig. 2.4) (see Chapter 9 for additional images) (7). It is likely that trigonal squamous mucosa in women represents a normal histologic variant unassociated to local injury. Under other pathologic conditions, the metaplastic squamous epithelium, that is, nonkeratinizing squamous mucosa in males and/or in nontrigonal locations or mucosa with keratinization in either sexes, may exhibit parakeratosis and even a
granular layer (8) (Fig. 2.5) (efig 2.13). This metaplastic epithelium is not preneoplastic per se but under some circumstances may lead to squamous carcinoma (see Chapter 9), particularly where multifocal and extensive.

The urothelium is markedly thickened and, importantly, lacks cytologic atypia (Fig. 2.6) (efigs 2.14-2.16). Rather than requiring a specific number of cell layers, marked thickening and/or increased cell density within the urothelium (increased cells per unit area) is needed to diagnose flat hyperplasia. This lesion may be seen in the flat mucosa adjacent to low-grade
papillary urothelial lesions. When seen alone in a de novo setting, there are no data to suggest that it has a premalignant potential. When these lesions occur in the setting of papillary tumors, flat hyperplasia may contain genetic abnormalities (chromosome 9 abnormalities and FGFR3 mutations) indicating that this may be an early manifestation of low-grade urothelial neoplasia (9, 10).

As “reactive urothelial atypia” may in some instances lead to confusion with dysplasia or cause concern for patients, some of the current authors have substituted the term “reactive urothelial changes.” Nucleomegaly is the most prominent finding in reactive urothelial changes, but the cells often have a single prominent nucleolus and evenly distributed vesicular chromatin (Fig. 2.7) (efigs 2.17-2.28). The nuclear borders are smooth. The nuclei are frequently round, and nuclear pleomorphism is lacking (11).

Architecturally, the cells maintain their polarity perpendicular to the basement membrane although a minimal loss of polarity may be evident. The mitotic rate may be increased with mitoses present predominantly in the basal and intermediate urothelium, but atypical forms are not seen. Intraurothelial acute or chronic inflammation is commonly identified (Figs. 2.8, 2.9, 2.10). It is important to recognize that even intraepithelial lymphocytes, by themselves, can result in reactive changes. The cytoplasm may become more basophilic or eosinophilic with loss of cytoplasmic clearing. Clinical history of stones, infection, or frequent instrumentation may be present (12). Reactive urothelium may be denuded with only a single residual layer of basal cells remaining; the residual cells are not hyperchromatic, not enlarged, and do not possess nuclear membrane irregularity.

The umbrella cells frequently undergo reactive change alone and exhibit multinucleation or nucleomegaly, often with cytoplasmic vacuolation. Although reactive conditions share the presence of enlarged nuclei, prominent nucleoli, and frequent mitoses with CIS, in reactive conditions, nuclei are uniform in size and shape with vesicular nuclei as opposed to the often hyperchromatic nuclei of CIS. Furthermore, most cases of CIS are not associated with intraepithelial inflammation.

This term was originally proposed as a descriptive category in cases with inflammation in which the severity of atypia appears to be out of proportion to the extent of inflammation such that dysplasia or CIS could not be confidently excluded (Figs. 2.11, 2.12) (efigs 2.29). The message
conveyed to the urologist by use of this diagnostic terminology was that the patient should be followed up after inflammation subsides. This was not a disease or diagnostic entity but merely a descriptive term that may be designated in a biopsy in which the flat lesion with atypia cannot be diagnosed with certainty into reactive, dysplastic, or CIS categories. Since this term is often used loosely and potentially as a “wastebasket” term by pathologists for lesions with minimal atypia that may well fall into the spectrum of “normal,” we dissuade the use of this term and suggest that pathologists acknowledge the level of diagnostic difficulty in cases where it is not possible to reliably exclude reactive from neoplastic atypia, than use this term. Studies have shown that cases diagnosed as reactive atypia and atypia of unknown significance have similar clinical features and are not associated with progression to bladder cancer (13, 14), further supporting our position.

Distinct appreciable nuclear abnormalities in the urothelium that occur in the absence of inflammation or that appear disproportionate to the amount of inflammation (if present), but that are not severe (i.e., falling below the threshold of CIS), may be designated as urothelial dysplasia/low-grade intraurothelial neoplasia (Figs. 2.13, 2.14, 2.15, 2.16) (efigs 2.20-2.39). It has been proposed that bladder intraepithelial lesions, like intraepithelial lesions of the cervix, be graded on the basis of level of involvement of atypical cells (i.e., mild dysplasia for lesions showing atypia confined to the lower one third, moderate dysplasia for atypia up to the middle third, and so on) (15). However, observations in animal models and in clinical specimens indicate the importance of cytologic features (degree of cytological atypia) rather than histologic pattern (extent of atypia from the basal layer) (16).

In dysplasia, the thickness of the urothelium is usually normal (three to seven layers) but may be decreased or increased (Fig. 2.17). Flat lesions with a benign cytology and minimal disorder should be considered within the spectrum of “normal” (17). Dysplastic lesions show loss of polarity (normal cells are columnar to oval with nuclei perpendicular to the basement membrane) evidenced by crowding of nuclei. The cells are more rounded to polygonal cells with nuclei parallel to the long axis. Nuclear atypia is evident but is not severe enough to merit a diagnosis of CIS. Often, increased cytoplasmic eosinophilia, nucleomegaly, irregularity of nuclear contours with notching of cell borders, and altered chromatin distribution are seen. Nucleoli are usually not conspicuous or are only randomly
present in the mucosa but are not typically present throughout. Only a minimal degree of pleomorphism is allowable in dysplasia, and the mitotic activity is variable although usually not in the higher layers. The lamina propria is usually unaltered, but it may contain increased inflammation or neovascularity. Denudation with atypical cells clinging to the submucosa is not a common feature of dysplasia. Comparison with more normalappearing urothelium, if present in the same biopsy specimen or in another simultaneously obtained biopsy specimen from the same patient, may help in assessing features such as nucleomegaly, loss of clearing, and loss of polarity. In some patients on surveillance for low-grade papillary urothelial carcinoma, due to increased frequency of cystoscopy, early lesions may be found, which histologically show thickened urothelium with dysplasia. Early papillary formation or neovascularity at base, similar to papillary hyperplasia, may be present. Similar abnormalities in the urothelium may be seen in the mucosa immediately adjacent to a low-grade papillary urothelial carcinoma (shoulder effect). Another way to conceptualize the difference between dysplasia and CIS is that the cytological findings seen in dysplasia are analogous to those seen in noninvasive low-grade papillary urothelial carcinoma, whereas CIS is analogous in its histology to noninvasive high-grade papillary urothelial carcinoma. Immunostaining is not recommended in the distinction between dysplasia and CIS (for potentially useful immunoprofile, see section on CIS below) as the distinction is based purely on morphologic grounds with the former denoting distinct nuclear atypia falling short of the threshold to be designated as CIS.

A diagnosis of primary dysplasia (i.e., no prior history or concomitant urothelial neoplasia) should be made with great caution (18, 19). Dysplasia is usually a histologic diagnosis seen most commonly in patients with bladder neoplasia, in whom the incidence is 22% to 86%. The incidence approaches 100% in patients with invasive carcinoma. Little is known about de novo dysplasia because there is a lack of screening in the general population. Furthermore, de novo dysplasia is usually cystoscopically and clinically silent, although it may appear as a slightly raised and mildly erythematous patch. The patients are predominantly middle-aged men presenting occasionally with irritative bladder symptoms with or without hematuria. The few studies published on de novo dysplasia indicate that 5% to 19% of patients progress to have urothelial neoplasia (20, 21, 22, 23). The finding of dysplasia in patients with bladder cancer suggests that it is a marker for progression (increased recurrences or future invasion) (13, 24, 25). However, the definition of dysplasia used in all these studies has been variable, so their true clinical meaning remains in doubt. In general, the diagnosis of dysplasia suffers from poor interobserver reproducibility, even among expert genitourinary pathologists.