This review summarizes published data dealing with the prevalence of high-g
rade prostatic intraepithelial neoplasia (HGPIN) in a variety of prostate t
issue samples. Additionally, we have attempted to document the relationship
between HGPIN and the pathological parameters of prostate cancer in autops
y and radical prostatectomy specimens. Studies reporting the prevalence of
HGPIN in needle biopsies, transurethral resection specimens and radical pro
statectomy specimens, and those documenting the lesion in postmortem settin
gs are compared. We also summarize studies in which the distribution and/or
extent of HGPIN was correlated with prostate cancer stage, grade and volum
e. There is significant variation in the reported frequency of HGPIN, parti
cularly in needle biopsy specimens, with a range of 0.8-23.9%. The factors
responsible for these discrepancies include the population studied, the lim
ited sample size that needle biopsies represent, diagnostic inconsistencies
and, possibly, tissue preparation/staining variables. Because of the impor
tant implications a diagnosis of HGPIN carries, there is a pressing need to
achieve greater consistency in diagnosing and reporting the lesion. Better
targeted educational efforts, including teaching courses, websites with il
lustrations and the possibility of teleconsultations, are among possible me
ans to attain this goal. Better documentation of the evolution of HGPIN to
cancer through clinical follow-up is also recommended.