Three Part Question
In [a male adult with fever] does [serum prostate specific antigen level] assist in [the diagnosis of acute prostatitis]?
Clinical Scenario
A 60-year-old man with fever for 5 days is brought to the emergency department by his family. He has no other somatic symptoms except for fever. The results of the physical examination are within normal limits except for a moderately enlarged prostate. Laboratory studies including chest film and urine analysis show no specific findings. You wonder whether serum prostate specific antigen level would assist in the diagnosis of acute prostatitis.
Search Strategy
Medline, OVID interface on the world wide web, 1966 to February 2008.
Cochrane Library on the world wide web
Nedline: {(exp prostate specific antigen OR PSA.mp) AND (exp prostatitis OR prostatitis.mp) AND (acute.mp)} LIMIT to human AND English.
Cochrane Library: [prostate]
Search Outcome
Medline: 43 papers were found; 3 were considered to be original research of high quality and relevant to the topic of interest (Table 1). No additional relevant citations were found in the Cochrane Library.
Relevant Paper(s)
Author, date and country |
Patient group |
Study type (level of evidence) |
Outcomes |
Key results |
Study Weaknesses |
Hara et al 2004 Japan | 39 patients (age 24-79 years) with fever of unknown origin or symptoms of acute pyelonephritis | Cohort study | PSA levels | For PSA level >4 ng/mL, acute prostatitis (n=20) was predicted with sensitivity:100%, specificity: 100%, PPV: 100%, NPV:100%. | Small number of patients
Unclear blinding process and reference standard for diagnosis |
Pansadoro et al 1996 Italy | 72 patients (age < 50 years) with symptoms of prostatitis | Cohort study | PSA levels | For PSA level >4 ng/mL, acute prostatitis (n=7) was predicted with sensitivity:71%, specificity:94%, PPV:56%, NPV:97%, LR+:11.6. | Small number of patients
It is unclear whether patients with fever were included. |
Morote et al 1986 Spain | 996 male patients (age 2-64 years) without malignant prostatic pathology | Cohort study | PSA levels | For PSA level >4 ng/mL, acute prostatitis (n=50) was predicted with sensitivity: 52%, specificity: 77%, PPV: 11%, NPV:97%, LR+:2.3. | Small number of patients
It is unclear whether patients with fever were included.
Unclear reference standard for diagnosis |
Comment(s)
The clinical presentation of acute prostatitis is variable among patients, including fever, chills, dysuria, urgency, frequency, and tender prostate. Urine analysis usually showed pyuria. Nevertheless, fever may be the only symptom and dilemma in the diagnosis of acute prostatitis occurs sometimes. Serum PSA level has been proven to be elevated in prostate cancer, benign prostate hyperplasia, and prostatitis. However, the role of serum PSA level in the diagnosis of acute prostatitis is still unclear. The studies in Table 1 illustrate that the use of PSA may be of some help in the diagnosis of acute prostatitis. However, all of these studies suffer from quality issues, such as unclear blinding process, unclear reference standards, and small number of patients. Although all three studies have a very good negative predictive value (from 96.83% to 100%), this should be considered carefully due to the relatively small case numbers and lack of complete final confirmation of acute prostatitis.
Clinical Bottom Line
There is insufficient evidence to support the use of serum prostate specific antigen in the diagnosis of acute prostatitis. However, the negative predictive value is high. Further research is needed to answer this question.
References
- Hara N, Koike H, Ogino S, et al. Application of serum PSA to identify acute bacterial prostatitis in patients with fever of unknown origin or symptoms of acute pyelonephritis. Prostate 2004;60:282-288
- Pansadoro V, Emiliozzi P, Defidio L, et al. Prostate-specific antigen and prostatitis in man under fifty. Eur Urol 1996;30:24-27
- Morote J, Ruibal A, Palou J. Evaluation of specific antigen and prostatic acid phosphatase specificity. Study of false values. Int J Biol Markers 1986;3:141-146