Fehlerkriterien bei der PSA-Diagnostik

Abstract

The widespread use of prostate-specific antigen (PSA) determination in the diagnosis of prostate cancer has proved to be generally beneficial; however, as a result expert commissions, arbitration committees and the courts have had to deal with an increased number of suspected treatment errors. As a follow-up to the previous report on the decisions made by expert commissions, this paper deals with recent developments and their assessment. The procedures followed for assessment have been extensively described in the previous paper. The criteria for assessment of disputed treatment were and are the accepted standards (i.e. the standards applicable to medical specialists) and the quality of care applied in accordance with the pertinent definitions. In the period from 2005 to 2011 (i.e. 7 years) errors in medical treatment were determined in connection with PSA determinations in 22 out of the 37 cases reviewed, i.e. 71%. These were subdivided into 3 cases from general practitioners, 5 cases from specialists in internal medicine and 15 cases from urologists (in 1 case 2 different doctors were involved). They were faulted for omitting a follow-up biopsy of the prostate. In 12 cases this involved PSA values above the recommended cut-off level without suspicious palpation results, in 7 cases raised PSA levels with suspicious palpation findings, in 2 cases suspicious palpation findings without raised PSA and in 1 case the omission of both palpation and PSA determination. An error in treatment was negated if the PSA value was below the recommended cut-off value or had fallen below it again subsequently (two cases each), if follow-up prostate biopsy was recommended and documented following the determination of raised PSA and/or suspicious palpation findings (three cases) or if follow-up treatment was rejected in spite of a documented recommendation (one case). Treatment errors in association with PSA determinations can therefore be uniformly and plausibly assessed using objective criteria and can thus be avoided.