International Journal of STD & AIDS 1997; 8: 475± 481

EDITORIAL REVIEW

Diagnosis of chronic prostatitis: overview and update R N Thin MD FRCP Department of Genitourinary Medicine, St Thomas’ Hospital, Guys and St Thomas Hospital Trust, London, UK Summary: Symptoms and signs are unhelpful in the diagnosis of chronic prostatitis which in many cases continues to rest on comparison of white cells and organisms in urine samples collected before (VB2) and after (VB3) prostatic massage to express prostatic secretion (EPS), and particularly in the EPS itself, if this is obtained. A series of 195 patients is reviewed, 38 with chronic bacterial prostatitis (CBP), 66 with chronic non-bacterial prostatitis (CNBP), 55 with prostatodynia, and 31 with a history of recurrent urethritis without prostatitis. Demographic characteristics and history of recurrent urethritis were similar in all groups indicating that recurrent urethritis alone does not predispose to prostatitis. The upper limit of normal for the EPS while cell count was taken as 1000/mm3 in line with other reports. With this, the upper limit of normal for the estimate of white cells by simple microscopy appeared to be about 5/high power ® eld (hpf) rather than the ® gure of 10 often quoted; with the latter ® gure, a number of cases of CNBP would have been missed. All microscopy was undertaken with the same microscope using a 40 objective. Culture results showed a predominance of enterococci, and cultural and cytological ® ndings in EPS and VB3 were comparable. On microscopy, clumping of white cells was associated with increased numbersÐ mentioned previously in the literature but not supported by data. Ejaculation just before examination was associated with reduced rather than the increased numbers of cells previously reported. Individual investigators should assess their own methods in determining upper limits of normal for cells. In a separate series of 8 patients with symptoms compatible with prostatitis, transrectal ultrasound scanning showed a prostatic cyst; aspiration was associated with relief of symptoms. It is concluded that transrectal ultrasound scanning (TRUS) should precede prostatic investigation by prostatic massage as this may save the prolonged treatment often necessary for prostatitis.

´

Keywords: Chronic bacterial and non-bacterial prostatitis, prostatodynia, diagnosis

INTRODUCTION

Anatomy

Chronic prostatitis continues to be regarded as an obscure, ill-de® ned condition. In 1991 it was suggested that this was due to various factors including the anatomical situation of the organÐ deeply placed and inaccessible to clinical examination, no distinguishing clinical features, and the dif® culties in investigation1. Confusion persists regarding aetiology and diagnosis2. A further series of patients has been assessed since the report of Thin and Simmons in 19833. In this overview, the principles of diagnosis will be summarized, and the demographic and diagnostic features of the new series presented to help understand the condition.

The prostate is a tuboalveolar gland with a ® bromuscular stroma and 4 glandular zones within a ® brous capsule. The peripheral zone is the largest and with the central zone comprises much of the prostatic volume. Between these zones, in young men, lies a narrow transitional zone, and there is a thin periurethral zone. These zones have pathogenic importance. The transitional zone is the site of benign prostatic hypertrophy in older men, while in¯ ammation and cancer develop in the peripheral zone4.

Correspondence to: Dr R N Thin, Department of Genitourinary Medicine, St Thomas’ Hospital, London SE1 7EH, UK

DIAGNOSIS Symptoms are not diagnostic. Digital examination is frequently normal with 2 exceptions; localized tenderness indicates prostatitis, while generalized

475

476

International Journal of STD & AIDS

Volume 8

August 1997

prostatic tenderness suggests prostatitis but may be dif® cult to interpret. Tenderness contraindicates prostatic massage. Diagnosis usually depends on investigations. Urine and prostatic ¯ uid examination Accepted methods of diagnosis include cytological and cultural examination of urine passed before and after prostatic massage, and examination of the EPS5. The numbers of polymorphonuclear leucocytes (PMNLs) which are normal or abnormal in the EPS are still not entirely clear and will be discussed. For patients attending GU Medicine/STD clinics, exclusion of urethritis plus urine culture to exclude bacteruria should precede prostatic massage6. These investigations will allow urethritis and bacteruria to be identi® ed and treated before prostatic massage, minimizing diagnostic confusion and complications of massage. The patient should undertake a 2-glass urine test immediately before prostatic massage; send the second sample (midstream urine or voided bladder urine 2 (VB2) to the laboratory to compare with the post prostatic massage urine or voided bladder urine 3 (VB3)5. Examination therefore includes VB2, prostatic massage to obtain EPS, and VB3. The PMNLs are counted in the fresh sample of EPS using a Fuchs Rosenthal counting chamber. PMNLs may also be assessed per high power ® eld (hpf) by simple microscopy3,6,7. Quantitative culture should be undertaken on all samples7. Ultrasound scanning Transrectal ultrasound scanning of the prostate is established and typical, though not diagnostic, features of prostatitis have been described 8. TRUS can reveal prostatic cysts, abscesses and disease of the seminal vesicles which may produce clinical features suggesting prostatitis8. TRUS may show calculi, bladder neck dysfunction, benign prostatic hypertrophy and carcinoma which can be confused or may co-exist with chronic prostatitis8,9. Therefore it is important to undertake TRUS before prostatic massage. TRUS may also indicate the rare granulomatous, eosinophilic and tuberculous forms of chronic prostatitis10. Though a normal scan may occur with abnormal EPS, a normal TRUS and EPS will help to re-assure the patient with prostatodynia. A major role of TRUS is in the management of focal disease to guide a needle for aspiration or biopsy11. Other investigations Raised pH in the EPS has been reported but has limited diagnostic value in individual cases1. Increase in prostatic ¯ uid immunoglobulins and prostate speci® c antigen have been noted. Reduced EPS speci® c gravity, prostate antibacterial factor, zinc, magnesium, calcium, citric acid, acid phosphatase and lysozyme have also been re-

ported. Prostatitis and prostatodynia produce qualitative changes in the proteins of the EPS11,12. These changes have limited diagnostic value. Sexually transmitted infections Though Chlamydia trachomatis has been implicated in the aetiology of chronic prostatitis, percutaneous prostatic aspiration has so far failed to demonstrate the organism. Studies which showed chlamydia did not distinguish between contamination and the prostatic origin of the organism. Views vary regarding antichlamydial antibody studies. Some regard them as inconclusive11 and others as suggesting an aetiological role for chlamydia in prostatitis13,14. A causative role for Ureaplasma urealyticum has been suggested but not established11. Neisseria gonorrhoeae is rarely if ever isolated in cases of chronic prostatitis diagnosed in developed countries, but in the past was reported in prostatic abscesses15,16. Trichomonas vaginalis may cause prostatitis17,18. Cultures may be more sensitive than simple microscopy of EPS or VB3 in identifying the parasite in secretions from men 18,19. However, T. vaginalis has not been reported in needle aspirates from the prostate. Classi® cation and de® nitions Classi® cation of chronic prostatitis remains as previously described, CBP, CNBP, and prostatodynia1. Chronic bacterial prostatitis is diagnosed when there is a pure culture of a urinary tract pathogen in EPS and/or VB3 with negative results in VB2. Chronic non-bacterial prostatitis is diagnosed when there are excess numbers of PMNLs in EPS and VB3 with normal numbers in VB2, and negative culture results in all 3 samples. Prostatodynia is the name given when there are clinical features of prostatitis such as perineal ache or pain on ejaculation, but normal numbers of cells and negative culture results in all samples. ANALYSIS OF NEW SERIES Patients and methods Since the report in 19833, an additional 195 patients have been studied giving a combined total of 358 patients (Table 1). Most of the following analyses apply to the new series. Patients were referred to a specialist clinic between 1983 and 1992 at St Peters’ Hospital, London, either because prostatitis was suspected or because of recurrent urethritis. Diagnostic methods were the same in both series and were as described above; cultures for chlamydia, mycoplasma and ureaplasma were not undertaken. Patients with CBP had a pure growth in the EPS of a pathogen with at least 1 103 organisms per ml7. Those with CNBP had at least 1000 PMNLs/

´

Thin. Diagnosis of chronic prostatitis

FINDINGS

Table 1. Number of cases in new and old series3

Chronic bacterial prostatitis Chronic non-bacterial prostatitis Prostatodynia Recurrent urethritis Control subjects

New

Old

Total

38 66 55 36

46 46 29 14 28 163

84 112 84 50 28 358

195

mm3 in the EPS, and no pathogens on culture20. All men with prostatodynia had symptoms, less than 500 PMNLs/mm3 in the EPS and no pathogens on culture. Most patients had a history of recurrent urethritis. All symptomatic patients had had their problems for at least a year. A group of patients was also identi® ed with no evidence of prostatitis or prostatodynia, but with a history of at least 3 episodes of NSU as previously described 21 or gonococcal urethritis (Gram stain and culture positive) over the previous year. This group will be called recurrent urethritis. Patients with a history suggesting bacteruria or documented bacteruria, local or generalized tenderness of the prostate, or in whom EPS, VB2 and VB3 could not be obtained, were excluded. Control subjects attended a GU medicine clinic with no history of urethritis or bacteruria, no symptoms and no urethritis, bacteruria or prostatitis on investigation6. Statistical comparisons were made by the Chi square test, Normal test, paired t test and the Sign test.

Details of patients with CBP, chronic nonbacterial prostatitis (CNBP), prostatodynia and recurrent urethritis without chronic prostatitis Demographic details of patients are summarized in Table 2. On the whole the groups were similar. Patients with CBP were on average a little older, 41.1 years compared to 32.5± 33.9 years. A few more men with prostatitis were born in the UK than in the other 2 groups, and a few more with prostatodynia were born abroad. The history of urethritis was similar in all 4 groups (Table 3). The main symptoms among patients with prostatitis and prostatodynia are summarized in Table 4. Pain was the most common complaint, being more often perineal in CBP, penile tip in CNBP and in the scrotum in prostatodynia. In addition 6 patients with CNBP and 5 with prostatodynia but none with CBP complained of pain on ejaculation. Patients with prostatodynia had a wide variety of other complaints including iliac fossa discomfort, supra-pubic discomfort, low backache and general genital pain or discomfort. Dysuria was uncommon in all groups as Table 4 shows. No patient complained of frequency or symptoms suggesting haematuria. General symptoms were not systematically recorded but malaise and tiredness were common on presentation. Most prostates felt normal on digital examination. EPS white cells The mean PMNL cell count± standard deviation in patients with CBP was 3022± 1995.6, and with CNBP was 2956± 1852.6. Many reports on prostatitis are based on EPS white cell estimates so counts

Table 2. Demographic details

Age Range (years) Mean Born UK Elsewhere Married Single Separated/Divorced

Chronic bacterial prostatitis (n=38)

Chronic non-bacterial prostatitis (n=66)

Prostatodynia (n=55)

Recurrent urethritis (n=36)

20± 50 41.1

20± 48 32.9

20± 56 33.9

21± 58 32.5

30 8 15 22 1

49 17 24 38 4

33 22 16 36 3

24 12 12 22 2

Table 3. History of urethritis: mean number of episodes Chronic bacterial prostatitis Non-speci® c urethritis Gonococcal urethritis

4.8 0.12

477

Chronic non-bacterial prostatitis 3.5 0.14

Prostatodynia 3.3 0.22

Recurrent urethritis 4.2 0.4

478

International Journal of STD & AIDS

Volume 8

August 1997

Table 4. Symptoms

Number of patients with symptoms Number of symptoms Number of c/os of pain Common pains Perineal Penile tip Scrotal Dysuria

Chronic bacterial prostatitis

Chronic non-bacterial prostatitis

Prostatodynia

27/38 37 33

48/66 58 49

55/55 61 55

13 5 6 4

5 15 11 8

9 10 13 6

and estimates were compared. In Table 5 the upper limit for the normal estimate is taken as 5 PMNLs/ hpf as before3 and shows a correlation between count and estimate. Other reports take the upper limit for the cell estimate as 10/mm3 11. The ® ndings in Table 6 suggest that for the methods used here, a minimum estimate of 10 would have excluded 27 of 36 patients with counts of 1000/ mm3 or more. Clumps of PMNLs in the EPS may indicate prostatitis22. A clump was taken as a close group of at least 5 PMNLs. Table 7 shows correlation between clumps and high counts and estimates, Table 5. Express prostatic secretion (EPS) white cell counts and estimates (A)

and between an absence of clumps and low counts and estimates. Clumps were associated with a mean estimate of 8.6 PMNLs/hpf again suggesting that an upper limit of 10 cells/hpf is too high. If prostatic massage fails to produce EPS, diagnosis may rest on VB3. The ® ndings in Table 8 indicate a correlation between EPS PMNL counts and estimates, and cells in VB3. The day after sexual activity, increased white cells in the EPS have been reported23; Table 9 indicates the reverse with more cells in samples taken 5± 7 days after ejaculation compared to samples taken from the same men 1± 2 days after ejaculation. Sampling was repeated after an interval of 1± 2 weeks. Samples were easier to obtain after the longer interval.

Estimate (cells/hpf) Count (PMNLs/mm3 )

0± 5.0

5.1 and above

1000 and over 0± 999

81 229

49 13

Table 8. White cells in express prostatic secretion (EPS) and VB3

c2=63, P< 0.001; PMNLS=polymorphonuclear leucocytes;

VB3 cells/mm3

hpf=

high power ® eld

Table 6. Express prostatic secretion (EPS) white cell counts and estimates (B) Cell count> 1000/mm3 (n=36) mean=4115 Mean cell estimate=9.3/hpf Cell estimate> (n=9) median=10 Median cell count=5000

Table 7. Express prostatic secretion (EPS) white cells: clumps versus no clumps

Clumps (n=41) No clumps (n=74)

Estimate (mean) cells/hpf

Count (mean) PMNLs/mm3

8.6 1.8

3654 936

<
1000 < 999 EPS cell estimate > 5.1 < 5.0

>

20