- Top of page
Objective To develop a standardized histopathological classification system for chronic prostatitis (standardized description of prostatic inflammatory infiltrates) based on a literature review, extensive prospective evaluations in two recognized prostatitis research centres and widespread consensus of international urological centres identified as having major expertise or interest in chronic prostatitis.
Methods Relevant articles for review were identified by a Medline search undertaken by the Cochrane Review Group in Prostate Diseases and Urologic Malignancies, and cross-checking bibliographies of retrieved studies, reviews, book chapters and abstracts of the American Urological Association and International Prostatitis Collaborative Network Annual Meetings. Initial drafts were based on classification systems independently developed by the Prostatitis Research Centers at Queen's University in Canada and University of Washington in the USA. A collaborative draft was distributed to 20 urological/pathological clinical centres who participated in the North American Chronic Prostatitis Collaborative Research Network and First International Prostatitis Collaborative Network. A consensus classification system was then distributed to the participating panel for acceptance.
Results The literature review identified a reasonably consistent description of inflammatory infiltrate locations and patterns that were further incorporated into the draft based on the Queen's University and University of Washington proposals. Eighteen (90%) of the identified Prostatitis Centers participated in the revision of the draft and the final consensus process. The final consensus document classifies prostatic inflammation according to its extent and grade/severity in each tissue compartment (location).
Conclusion The consensus of the expert panel was that this classification system can be used in the evaluation of prostatic inflammation in prostate biopsies, transurethral resected prostate chips or prostatectomy specimens. A standardized accepted framework to describe histopathological prostate inflammation will prove useful in evaluating prostate disease.
- Top of page
Different medical disciplines have traditionally defined prostatitis using very different criteria. The pathologist has usually given little attention to prostatic inflammation unless it was particularly florid, primarily because the clinical importance of the observation remains largely undefined. For the pathologist, ‘prostatitis’ is defined as ‘more’ (no one has ever discussed how many is ‘more’) inflammatory cells within the prostatic parenchyma . The pathologist based the diagnosis on the cell type that comprised the infiltrate, i.e. neutrophilic, mononuclear (including lymphocytes, plasma cells and macrophages), eosinophilic or granulomatous prostatitis.
In contrast, the urologist defined ‘prostatitis’ as a clinical syndrome characterized by genitourinary discomfort or pain . This clinical syndrome is sometimes, but not invariably, associated with white blood cells and/or bacteria in prostate-specific specimens (expressed prostatic fluid, urine specimens after prostatic massage, or semen). The new National Institute of Health (NIH) classification system for chronic prostatitis/CPPS  recognises a further category of chronic prostatitis, that of asymptomatic inflammatory prostatitis (Category IV). In this category, prostatic inflammation may be associated with BPH, prostate cancer or infertility, or it may be present in patients with no identifiable prostate disease. The establishment of a standardized histopathological classification system of chronic prostatitis will permit comparison of future studies evaluating, describing or associating prostatic inflammation with other related lower genitourinary conditions.
The association of prostatic inflammation with chronic bacterial prostatitis, inflammatory CPPS (Category IIIA CPPS or chronic nonbacterial prostatitis) or non-inflammatory CPPS (Category IIIB or prostatodynia) has not yet been completely defined. In an early study analysing inflammation and needle-biopsy samples from 24 patients, an acute inflammatory infiltrate was located predominantly within the lumina of prostate glands in six patients while 12 had clusters of predominantly periglandular lymphocytes . In another study, half of 18 patients with chronic nonbacterial prostatitis had no inflammation, three had ‘moderate’ inflammation while no patients had ‘severe’ inflammation . In a study seeking infectious agents in chronic nonbacterial prostatitis using ultrasound-guided biopsy, 88% of 60 patients had a sparse infiltrate of chronic inflammatory cells . In a similar study of patients with chronic bacterial prostatitis, Nickel et al. noted a lack of association of severe prostatic inflammation with positive bacterial cultures. In one of the first studies to systematically characterize inflammation of the prostate in patients who had satisfied strictly defined clinical laboratory criteria for chronic prostatitis/CPPS, True et al. found that the degree of inflammation was minimal to absent in > 95% of cases in which prostate biopsies were examined. This group evaluated prostatic inflammation for location (glandular, periglandular, stromal), extent (focal, multifocal, diffuse) and density of inflammatory cells (inflammatory cells/mm2). They also found no correlation between leukocytes in the expressed prostatic secretion and in the prostate in this population. Further work is needed to define the association between lower genitourinary symptoms, histopathological evidence of chronic inflammation and inflammatory infiltrates in prostate-specific specimens (expressed prostatic secretions, and urine and semen after prostatic massage).
Histological evidence of prostatic inflammation is almost universally found in association with BPH . Kohnen and Drach  found inflammatory infiltrates in 98% of prostates that were resected for BPH. The location of inflammation, which did not correlate significantly with presence or absence of a ‘significant infection’, was described as glandular, periglandular and/or stromal. Others  have confirmed the histological association of prostatic inflammation and BPH, while others [18,19] have characterized some of the specific inflammatory cells associated with BPH. Anim et al. attempted to locate the different types of inflammatory cells in prostatic lesions associated with BPH to determine the sequence of events in the cellular reaction. This study suggested that macrophages respond first to local injury or stimuli (at the prostatic glandular epithelial level), followed by recruitment of T lymphocytes which participate in the inflammatory response and accumulate around the damaged glands. B cell activity appears to be a late event. Irani et al. analysed the influence of prostatic inflammation on serum PSA concentrations in patients with exclusively BPH tissue on prostate biopsies. Inflammatory infiltration was graded as 0 (no inflammatory cells), 1 (scattered inflammatory cell infiltrate within the stroma with no lymphoid nodules), 2 (non-confluent lymphoid nodules) and 3 (large inflammatory areas with confluence of infiltrate). Inflammatory aggressiveness was graded as 0–3 by the degree of contact between the inflammatory cells and glandular epithelium. Grading did not include the types of inflammatory cells. Nickel et al. examined histological sections from prospectively collected TURP specimens from 80 patients with a diagnosis of BPH but no history or symptoms of prostatitis, using an immunostain for leukocyte-common antigen and scanned using a computerized image analysis system. Inflammation was identified in all cases, but the tissue surface areas involved averaged only 1.1% of the total specimen, with periglandular inflammation being the predominant pattern (0.5%). Foci of inflammation were categorized as glandular, periglandular or stromal, and inflammatory cell density graded from 0 to 3 using a scheme similar to that of Irani et al.. There was no significant difference between any combination of inflammation pattern, volume or grade of inflammation in catheterized vs uncatheterized patients, or culture-positive vs culture-negative cases. Neither total PSA nor PSA density showed a significant correlation with the extent, grade or location of inflammation. Atan et al. found no statistical correlations between irritative symptom score and the degree morphological evidence of prostatic inflammation in patients undergoing surgery for BPH. Anjum et al. found that there was a higher incidence of a glandular pattern of inflammation in patients with BPH who presented in acute urinary retention than a stromal pattern in a control group of patients with BPH undergoing TURP.
Several studies examined the association of prostatic inflammation and prostate cancer [24–26], but there was no clear relationship between these pathological entities. However, Irani et al. recently noted that patients with high-grade inflammation surrounding malignant glands in radical prostatectomy specimens had significantly more postoperative biochemical recurrence than patients with low-grade inflammation. However, prostatic inflammation can sometimes confound the detection of prostate cancer through its potential effects on PSA levels. Prostatitis, and more specifically bacterial prostatitis, was first shown to influence PSA levels shortly after assay kits were released for general use . Adequate treatment with appropriate antibiotics for acute prostatitis usually results in subsequent normalization of serum PSA levels. A clinical investigation revealed an elevated PSA level in 71% of patients with acute prostatitis, 15% of patients with chronic prostatitis and 6% with nonbacterial prostatitis, but in none with prostatodynia . In patients with bacterial prostatitis, the serum PSA level decreased to normal after antimicrobial therapy in most cases in the series. In 700 men who underwent prostatic needle biopsy for an elevated PSA or an abnormal DRE, 27% had histological evidence of prostatitis . Most (94%) had chronic inflammation, 6% had acute inflammation and 0.2% had granulomatous changes. In another study, serum PSA values significantly correlated with the extent of histologically confirmed acute and chronic active prostatitis, whereas there was no correlation between PSA and chronic inactive prostatitis . In that study acute prostatitis was defined as neutrophilic infiltrate in the glands, whereas chronic inactive prostatitis indicated only chronic inflammatory infiltrate in the glands and stroma. Nadler et al. compared the effects of histological evidence of prostatic inflammation in a larger screening population of men with elevated PSA concentrations to men with normal serum PSA concentrations who were biopsied because of a suspicious DRE. They reported that prostatic inflammation may be an important factor contributing to PSA elevation in men with no prostate cancer. Inflammation in prostatic tissue was characterized as acute (polymorphonuclear leukocytes within glandular or ductal lumina, their epithelium and/or adjacent stroma) and chronic (mononuclear cell infiltrate in the stoma around prostatic glands) prostatitis. They further graded inflammation on a 3-point scale of 0 (none), 1 (low grade) and 2 (high grade).
Men with infertility frequently (up to 22%) have evidence of an inflammatory process of the reproductive tract (usually detected by the high concentrations of leukocytes in the semen), yet bacterial cultures fail to reveal any pathological organisms . Several studies have shown that men with clinical prostatitis appear to have altered semen values, with more leukocytes in the semen, decreased sperm motility and a decrease in the percentage of sperm with abnormal forms; this may be reversed with antibiotic therapy [33–39]. Biochemical data indicate that the prostate is the main target in clinically ‘silent’ male genital tract inflammation in the infertile male population . However, there are no published studies that correlate inflammatory cells in semen with tissue inflammation.
NIH Category IV (asymptomatic inflammatory prostatitis) is defined as the presence of demonstrable prostatic inflammation in prostate-specific specimens, including biopsies. Inflammation in autopsy series, which includes men with and without specific prostate disease, is reported to be 5–15.3% in patients over 60 years of age . McNeal  found inflammation in 40 of 90 (44%) sampled adult prostates obtained at autopsy. Evidence of prostatic inflammation was noted in 45% of aspiration biopsy specimens taken because of suspected carcinoma . Blumenfeld et al. noted a significant incidence of incidental lymphocytic prostatitis (usually stromal) in benign prostate glands. The relevance of prostatic inflammation in asymptomatic men, particularly those with no evidence of BPH or prostate cancer, is unknown.
A consistent description of several fairly distinct, although often coexisting, patterns of chronic inflammation has emerged from the present review of the literature. The most common pattern of inflammation is a lymphocytic infiltrate in the stroma immediately next to the prostatic acini, with a less prominent intraepithelial component [7,9,16]. This periglandular inflammation is typically patchy and although all anatomical regions of the prostate can be involved, there appears to be a predilection for the peripheral zone [45,46]. The intensity of the inflammatory process varies considerably, from only scattered lymphocytes to dense lymphoid nodules or secondary follicles with germinal centres. In cases in which there are many periglandular inflammatory cells, the degree of epithelial infiltration is typically also greater and effects on the epithelial cells are more prominent (although the basal cell appears more resistant to inflammatory damage) . Increased proliferation can be shown in an inflamed glandular epithelium by radiolabelled thymidine uptake and immunohistochemical staining with proliferation markers such as Ki-67 . The inflammatory infiltrates of the periglandular area usually consist of a mixed population of B and T lymphocytes, with greater overall numbers of T-cells [47,49,50]. Significant numbers of macrophage are also typically present both in glands and in the periglandular stroma, but plasma cells are less common.
Stromal lymphocytic infiltrates frequently coexist with periglandular inflammation. Sheets, clusters and occasional nodules of lymphocytes and scattered plasma cells are seen within the fibromuscular stroma, with no apparent relationship to glands. The tissue response to the infiltrates is usually minimal and significant stromal fibrosis is virtually always absent [7,16].
Infiltrates of inflammatory cells restricted to the glandular epithelium and lumen are found in association with BPH, prostatitis and as an isolated finding in asymptomatic patients [7,9,16]. The intraepithelial inflammatory cells may be lymphocytes, neutrophils and macrophages, while neutrophils and macrophages are typically found in the lumen. Most commonly, these patients show focal mild acute (i.e. neutrophil-rich) glandular inflammation combined with chronic periglandular and stromal inflammatory infiltrates [7,9,16].
Several other histological findings are not incorporated in the present consensus classification system. Corpora amylacea, which are commonly present within prostate glands and ducts, are not usually associated with prostatic inflammation. They may become large enough to distend a duct or gland, resulting in the attenuation of the adjacent epithelium, causing periglandular inflammation . Calculi usually occur in the central prostatic ducts and are not usually associated with inflammation. These prostatic calculi may sometimes contribute to prostatic inflammation by providing a nidus in which infectious organisms may escape host defences, and by preventing drainage through obstruction of the central ducts [51,52]. Peri-urethritis can also accompany chronic prostatic inflammation and is characterized by mixed populations of inflammatory cells, including lymphocytes, plasma cells and neutrophils, and is often associated with prominent oedema . Interpretation of the significance of these latter observations is difficult and by consensus it was decided not to incorporate these entities into the classification system.
The expert panel was not able to reach a consensus on the classification of granulomatous inflammation. The histological pattern of nonspecific granulomatous prostatitis is variable. It is typified by heavy lobular, mixed, inflammatory infiltrates which include abundant histiocytes, lymphocytes and plasma cells. The inflamed areas are often punctuated by disrupted prostatic ducts. Although occasional small, discreet granulomas may be present, most of the infiltrate is usually not organized into well-defined granulomas. There may be scattered small foci of necrosis and neutrophils, and even eosinophils can be prominent [53–57]. Granulomatous prostatic inflammation is a common complication of local prostate surgery [58,59], BCG therapy [60–62], recent UTI  or a rare event in a setting of systemic tuberculosis [64,65]. Granulomatous prostatitis can also, rarely, result from cryptococci, sarcoid, malacoplakia, and from an allergic vasculitis [66,67]. However, many cases of granulomatous prostatitis are idiopathic. By consensus, it was decided in this classification system to descriptively note the extent, location and severity of granulomatous inflammation.
The present histopathological classification system of chronic prostatitis is far from perfect. It is based on critical research published previously, particularly that of Kohnen and Drach , describing inflammatory patterns from prostate specimens in patients with BPH. The classification draws on the clinical and research experience of two prostatitis research centres that were independently developing classification systems of their own (Queen's University and University of Washington), and benefits from the consensus input of a panel of recognized experts (participants of the North American CPCRN and the IPCN). The system also, to some extent, suffers from these same associations; as with any classification system, a balance is required between simplicity (to allow for ease of use) and completeness. A document that was drafted according to such a wide consensus is bound to draw criticism from those who are more interested in one specific aspect of such a system. We have also failed to reach a consensus on the frequency, importance and classification of granulomatous inflammation, but by consensual agreement, this omission has been dealt with descriptively. Finally, this system will not be the final model on the histopathology of chronic inflammation of the prostate gland. As more research is undertaken, further evidence may require changes and/or modifications in the proposed system. However, there is no doubt that this proposed histopathological classification system of chronic prostatitis is an improvement over the status quo, i.e. no accepted classification system at all.
In conclusion, prostatic inflammation is a common histopathological observation; its association with chronic prostatitis/CPPS has not yet been completely defined. Even less is known about asymptomatic inflammatory prostatitis associated with BPH, prostate cancer and infertility. Inflammatory patterns seen in normal, asymptomatic prostate-specific disease-free men has not been adequately studied. There was a general consensus of urologists and pathologists with expertise and interest in prostatitis that a standardized accepted framework to describe histopathological prostatic inflammation will prove useful in evaluating these prostatic diseases.