The average age for the 30 patients was 51.33 years, (range, 26–76 years; SD 14.19). Symptoms most commonly recorded in the patients' charts were obstructive urinary symptoms (80%), pain (73%), sexual dysfunction (70%), and irritative urinary symptoms (60%) (Table 1).

Figure 1

Table 1: Symptoms recorded in the charts of 30 men with chronic urogenital complaints.

For all 30 patients, the average global symptom severity score dropped from a mean of 6.40 (range, 3-10; SD, 2.04) at the first visit to a mean of 1.95 (range, 0-9; SD, 2.25) at the last visit, a reduction of 69.53%, which was statistically significant (mean difference, 4.44; CI, 3.37 – 5.52); t(29) = 8.47; P < 0.0001) (Table 2).

Figure 2

Table 2: Symptom scores for 30 men. V = visit. First, peak, and last symptom

Symptom score peaks tended to occur at the beginning of treatment (Figure 1).

Figure 3

Figure 1: Symptom score peaks in 30 men. Twenty-nine of 30 men had their first symptom score peak at the first or second visit. All symptom score peaks are graphed.

Ten of the 30 patients (33%; mean age, 51.00 years; range, 26–72 years; SD 14.60) reported complete remission of their symptoms by their last clinic visit. Their average symptom score dropped from 6.60 (range, 3-10; SD, 2.59) at the beginning of treatment to 0.00 at the end, which was significant (mean difference, 6.60; CI, 4.75 – 8.45; t(9) = 8.06; P < .0001).

Twenty of the 30 men (67%; mean age, 51.50 years; range, 26–76 years; SD, 14.36) did not report complete remission of symptoms. In this group of 20 men, the mean first symptom score of 6.30 (range, 3–10; SD, 1.78) declined to a mean last symptom score of 2.93 (range, 0.10-9.00; SD 2.17), which was also significant (mean difference, 3.37; CI, 2.25 – 4.49; t(19) = 6.27; P < 0.0001).

Within this group of 20 men without complete remission of symptoms were 3 men (mean age, 47.33; range, 40–62 years; SD, 12.70) who reported not being any better at the end of treatment. Two of these 3 men reported the same symptom score at the beginning of treatment as at the end of treatment, and in 1 of the 3 men, the symptom score was higher at the end of therapy than it was at the start of therapy. In these 3 men, the mean symptom score rose from 6.67 (range, 6-8; SD, 1.15) at the beginning of treatment to 7.00 (range, 6-9; SD, 1.73) at the end of treatment.

Our graph shows the mean first symptom score, the mean peak symptom score, and the mean last symptom score of the 10 men reporting complete remission of symptoms versus the 20 men not reporting complete remission of symptoms (Figure 2).

Figure 4

Figure 2: Graph of the mean first, peak, and last symptom scores in 10 men reporting complete remission of symptoms versus 20 men not reporting complete remission of symptoms.

Smears of the urethral tip were collected 514 times in the 30 men (mean, 17.13; range, 10-35; SD, 4.80).

Bacteria were present in 28 of 30 (93%) of the men's Gram-stained urethral smears during treatment. By the last examination, bacteria were present in only 6 of the 30 (20%) Gram-stained urethral smears. Gram-positive bacteria were seen in urethral Gram-stains more often than were Gram-negative bacteria (Table 3).

Figure 5

Table 3: The presence of bacteria in the urethral Gram-stains of 30 men.

Urethral WBC counts declined to zero over time in all 30 men (Table 4).

Figure 6

Table 4: Urethral WBC counts, first, peak, and last, and all data points in 30 men.

We performed Friedman's test to test for differences between the first, peak, and last urethral WBC counts in the 30 men and the test was significant (P < 0.0001). The mean urethral WBC counts are tabulated (Table 5).

Figure 7

Table 5: Mean first, peak, and last urethral WBC counts in 30 men.

Using the Wilcoxon signed ranks test and our Bonferroni correction of less than 0.017 being significant, we then compared paired groups of data. The difference between the first urethral WBC counts (mean, 3.07; median, 0.00) and the peak (mean, 6.93; median, 1.50) was significant (P = 0.003). The difference between the peak (mean, 6.93; median, 1.50) and the last (mean, 0.00; median, 0.00) was significant (P < 0.0001). In addition, the difference between the first urethral WBC count (mean, 3.07; median 0.00) and the last urethral WBC count (mean, 0.00; median 0.00) was significant (P = 0.012).

Urethral WBC counts tended to peak early during therapy. Sixteen men had microscopic urethral purulence during therapy. Fourteen of the 17 (one man had two peaks) urethral WBC peaks occurred during the first 5 specimen collections (Figure 3).

Figure 8

Figure 3: All urethral WBC peaks are graphed (one patient had two peaks). Fourteen of the 17 (82.35%) urethral WBC peaks occurred during the patients' first 5 clinic visits.

We graphed the mean first, peak, and last urethral WBC counts of the 10 men who reported a complete remission of symptoms versus the 20 men who did not report a complete remission of symptoms (Figure 4).

Figure 9

Figure 4: Mean first, peak, last, urethral WBC counts in 10 men with complete remission of symptoms versus 20 men without complete remission of symptoms.

We also tabulated the mean and median urethral WBC counts in the 10 men who underwent a complete remission of symptoms versus the 20 men who did not undergo a complete remission of symptoms to demonstrate that the urethral fluid from the 10 men was more purulent at the first and peak specimens than that of the 20 men (Table 6).

Figure 10

Table 6. Mean and median first, peak, and last urethral WBC counts in 10 men with a complete remission of symptoms versus 20 men without complete remission of symptoms.

We used the Mann-Whitney test to compare the first urethral WBC counts in 10 men who underwent complete remission versus the first urethral WBC counts in 20 men who did not undergo complete remission, and the difference did not reach significance (P = 0.553). We compared the peak urethral WBC counts in 10 men versus the peak urethral WBC counts in 20 men, and the difference was significant (P = 0.015). There was no difference in the last values between the 2 groups as all values were zero.

The urethral smears were examined for RBCs, but they were seen only once during the 514 urethral smear examinations, at the first clinic visit, in patient No. 12. All urethral RBC counts were zero at all other times, which suggests that repetitive prostatic massage is not traumatic.

Eight of the 30 men had fluorescing C. trachomatis elementary bodies by (direct fluorescent antibody) DFA at their first visit. Eight more men showed fluorescing C. trachomatis elementary bodies at a later DFA test during the protocol; overall, 16 of the 30 men (53.33%) showed fluorescing C. trachomatis elementary bodies. Fifteen of the 16 Chlamydia positive men had fluorescing C. trachomatis elementary bodies seen during their first or second visit. Fourteen of the 16 (87.50%) men with fluorescing C. trachomatis elementary bodies had a negative C. trachomatis DFA recorded in their charts prior to discharge from the treatment protocol (Table 7).

Figure 11

Table 7: Number of fluorescing elementary bodies seen at each test that was performed. Only patient No. 8 did not undergo at least 1 Chlamydia DFA test.

Of the 16 men with fluorescing C. trachomatis elementary bodies, 10 (62.5%) had urethral WBCs at some point during the treatment period, while 6 (42.85%) did not display urethral purulence at any time during treatment.

The C. trachomatis DFA testing was not performed at standard intervals, so we make no strong conclusions. We noted; however, that in the 10 patients in this study who reported complete remission of symptoms, 9 had 1 or more fluorescing C. trachomatis elementary bodies, and in 8 of those 9 patients, a repeat DFA test documented that fluorescing C. trachomatis elementary bodies had disappeared. Overall, doing only 1 DFA test would have missed finding 7 patients with fluorescing Chlamydia elementary bodies.

Five patients (Nos. 2, 10, 11, 26, and 30) had yeast reported in their urethral smears at some point by microscopy.

Prostatic massage with expressed prostatic secretions (EPS) collection was performed 514 times in the 30 men (mean, 17.13; range, 10-35; SD, 4.80).

The Gram-stains of the EPS revealed visible bacteria by light microscopy in only 3 (10%) men (patients No. 12, 19, and 26). All three of these men also demonstrated fluorescing C. trachomatis elementary bodies (Table 8).

Figure 12

Table 8: EPS Gram-stain results for the 3 men who had findings are tabulated, along with first and last symptom scores. GNB = Gram-negative bacilli, GNCoB = Gram-negative coccobacilli, GPC = Gram-positive cocci, GPB = Gram-positive bacilli.

Patient No. 19, who had multiple bacteria seen by Gram-staining his EPS, only dropped in symptom score from 5 at the beginning of treatment to 4 at the end of treatment, while patients Nos. 12 and 26 responded well to treatment.

The mean EPS white blood cell counts for the 30 men demonstrated a rise and fall during treatment (Table 9).

Figure 13

Table 9: EPS WBC counts in the 30 men. First, peak, and last values are listed, followed by all data points.

Friedman's test revealed a significant difference between the first, peak, and last EPS WBC counts in the 30 men (P < 0.0001). Using the Wilcoxon signed ranks test with our Bonferroni correction of P less than 0.017 being significant, the first EPS WBC counts (mean, 12.40; median, 6.00) and the peak EPS WBC counts (mean, 36.50; median, 39.00) were significantly different (P = 0.0001). The peak EPS WBC counts (mean, 36.50; median, 39.00) and the last EPS WBC counts (mean, 10.53; median, 6.50) were also significantly different (P = 0.0001). The difference between the first and last EPS WBC groups did not reach significance (P = 0.810) (Table 10).

Figure 14

Table 10: The mean, range, standard deviation, and median first, peak, and last EPS WBC counts.

The rise and fall of EPS WBC counts demonstrates that the clinical horizon exists and suggests that obstructed prostatic acini are being drained during treatment.

When examining the EPS WBC counts under 1000X microscropy, 17 of the 30 (56.67%) men had 5 or more WBCs at their first massage, while 29 of the 30 men (96.67%) had an EPS WBC count of 5 or more by the eighth prostatic massage. Patient No. 8 was the only patient who did not have 5 or more WBCs in the EPS during the treatment period. Only 11 of the 30 men (36.67%) had an EPS WBC count over 10 at the first massage, while 26 of the 30 men (86.66%) had an EPS WBC count over 10 at some point during therapy. These results show that repetitive prostatic massage shifts most men from prostatodynia into the classification of inflammatory prostatitis.

Only 3 of 30 men (10%) had their most elevated EPS WBC count at their first prostatic massage, while 6 of the 30 men (20.00%) peaked at the sixth massage. It took 6 massages per patient to achieve the EPS WBC peak in over 50% of the patients. None of the EPS WBC counts peaked at the last EPS collection in any of the men studied. These findings have important implications, and clearly show that many prostatic massages must be done to obtain the most purulent EPS specimen for disease classification and microbial testing (Figure 5).

Figure 15

Figure 5: All EPS WBC peaks are graphed. The highest number of peaks was at the sixth prostatic massage. Thirty of 36 (83.33%) of peaks occurred during the first 10 visits.

We graphed the mean EPS WBC counts for the 10 men who reported a complete remission of symptoms versus the 20 men who did not report a complete remission of symptoms (Figure 6).

Figure 16

Figure 6: Mean EPS WBCs in 10 men versus 20 men at the first, peak, and last visits.

We also tabulated the mean and median first, peak, and last EPS WBC counts for the 10 men who reported a complete remission of symptoms versus the 20 men who did not report a complete remission of symptoms (Table 11).

Figure 17

Table 11: Mean with range and standard deviation, and median first, peak, and last EPS WBC counts in 10 men who reported complete remission of symptoms, versus 20 men who did not report complete remission of symptoms.

We used the Mann-Whitney test to determine if the first EPS WBC counts in 10 men differed from the first WBC counts in 20 men in a statistically significant fashion, and the difference did not reach significance (P = 0.102). The difference between the peak EPS WBC counts for the 10 men and the peak EPS WBCs counts for the 20 men also did not reach significance (P = 0.255). And finally, the difference between the last EPS WBC counts for 10 men and the last EPS WBC counts for 20 men did not reach significance (P = 0.508).

We compared mean EPS WBC counts at each visit with the urethral WBC counts at each visit, since both fluids were examined under 1000X magnification. The mean of the mean urethral WBC counts over 35 visits was 0.38 (range, 0.00–4.07; SD, 0.96; median, 0.00) and the mean of the mean EPS WBC counts over 35 visits was 14.26 (range, 1–28; SD, 5.12; median, 14.53). The difference between the mean EPS WBC counts and the mean urethral WBC counts at each visit was significant (P < 0.0001). The EPS was significantly more purulent than the urethral smears.

We compared the peak urethral WBC counts in the 16 men who had urethral WBCs present against the peak EPS WBC counts. The urethral WBC peaks tended to occur earlier than the EPS WBC peaks (Figure 7).

Figure 18

Figure 7: Urethral WBC peaks versus EPS WBC peaks. The peaks in urethral inflammation tended to occur earlier in therapy than EPS WBC peaks. All peaks are graphed.

We tabulated the EPS RBC counts in the 30 men, which went to zero over time in all 30 men (Table 12).

Figure 19

Table 12: EPS red blood cells, first, peak, last, and all data points for the 30 men. EPS RBCs went to zero by the last visit in all patients.

We performed Friedman's test to test for differences between the first, peak, and last EPS RBC counts, and a significant difference existed (P < 0.0001). We applied the Wilcoxon signed ranks test with our Bonferroni correction of P less than 0.017 as significant to compare paired groups. The difference between the first EPS RBC counts (mean, 0.17; range, 0–5; SD, 0.91; median, 0.00) and the peak EPS RBC counts (mean, 4.10; range, 0–45; SD, 9.46; median, 0.00) was significant (P = 0.001). The difference between the peak EPS RBC counts (mean, 4.10; median; 0.00) and the last EPS RBC counts (mean, 0.00; median 0.00) was significant (P = 0.001). The difference between the first EPS RBC counts (mean, 0.17; median 0.00) and the last EPS RBC counts (mean, 0.00, median 0.00) was not significant (P = 0.317), possibly due to small numbers.

We graphed the EPS WBC peaks and the EPS RBC peaks, and the peaks tended to overlap. All but one of the EPS RBC peaks occurred between the second to ninth visits, with the highest number of peaks occurring at the fifth and sixth massages (Figure 8).

Figure 20

Figure 8: All EPS WBC peaks versus all EPS RBC peaks.

We graphed the mean EPS RBC counts at the first, peak, and last visits for the 10 men reporting a complete remission of symptoms versus the 20 men not reporting a complete remission of symptoms (Figure 9).

Figure 21

Figure 9: Mean first, peak, and last EPS RBC counts in 10 men reporting a complete remission of symptoms versus 20 men not reporting a complete remission of symptoms.

We also tabulated the mean and median first, peak, and last RBC counts in the 10 men who reported a complete remission of symptoms to compare with the 20 men who did not report a complete remission of symptoms (Table 13).

Figure 22

Table 13: Mean, range, standard deviation, and median EPS WBC counts in 10 men reporting a complete remission of symptoms and the 20 men not reporting a complete remission of symptoms.

We used the Mann-Whitney test to compare the first EPS RBC counts in the 10 men versus the first EPS RBC counts in the 20 men and the difference was not significant (P = 0.48). The difference between the peak EPS RBC counts in the 10 men versus the peak EPS RBC counts in the 20 men also did not reach significance (P = 0.592). The last EPS RBC values were all zero in both groups.

Patient No. 11 was the only patient to have yeast seen in his EPS during treatment. He was one of the 10 patients who reported complete remission of symptoms.

All patients had a complete set of EPS cultures done on their first EPS specimen. EPS cultures were sometimes repeated during treatment, especially if the EPS WBC count rose during treatment. Staphylococcus species were most commonly cultured. One patient, patient No. 9, was positive for Ureaplasma (Table 14).

Figure 23

Table 14: EPS WBC culture results in 30 men with chronic urogenital complaints.

We compared the 5 men who were E. coli positive (patients Nos. 7, 11, 14, 25, and 26) with the 25 who were E. coli negative for descriptive purposes. We noted that the 5 men who were E. coli positive were less likely to have as great an improvement in symptom scores over time, were less likely to have urethral inflammation, had lower mean first EPS WBC values, had no EPS RBCs, and had lower peaks in post-massage urinary WBCs.

Post-massage urine samples were obtained 514 times in the 30 men (mean, 17.13; range, 10–35; SD, 4.80). We tabulated the urinary WBCs (Table 15).

Figure 24

Table 15: Urinary WBC counts, first, peak, last, and all data points for the 30 men.

We tabulated the mean and median values for the urinary WBC counts in the 30 men (Table 16).

Figure 25

Table 16: Mean, range, standard deviation, and median first, peak, and last urinary WBC counts in the 30 men.

The Friedman test was significant for differences between the first, last, and peak urinary WBC counts in the 30 men (P < .0001). We applied the Wilcoxon signed ranks test using our Bonferonni correction of P less than 0.017 being significant to compare the pairs of groups. The difference between the first urinary WBC counts (mean, 10.63; median, 4.00) and the peak urinary WBC counts (mean, 30.73; median 25.00) was significant (P < 0.0001). The difference between the peak urinary WBC counts (mean, 30.73; median, 25.00) and the last urinary WBC counts (mean, 6.93; median, 3.00) was also significant (P < .0001). The difference between the first urinary WBC counts (mean, 10.60; median, 4.00) and the last urinary WBC counts (mean, 6.93; median 3.00) did not reach significance (P = 0.855).

We graphed all the urinary WBC peaks in the 30 men along with the EPS WBC peaks, and our graph demonstrates a temporal relationship between the EPS and urinary WBC peaks, indicating that the post-massage urine purulence reflects the EPS (Figure 10).

Figure 26

Figure 10: EPS and Urinary WBC peaks in 30 men. All peaks are graphed.

We graphed the urinary WBC counts of the 10 men who reported complete remission of symptoms versus the 20 men who did not report complete remission of symptoms. The mean first and peak urinary WBC counts were higher in the men who underwent complete remission of symptoms (Figure 11).

Figure 27

Figure 11: Mean urine WBC counts, first, peak, and last, in 10 men with complete remission of symptoms versus 20 men without complete remission of symptoms.

We tabulated the mean and median urinary WBC counts for the 10 men who underwent a complete remission of symptoms as well as for the 20 men who did not report undergoing a complete remission of symptoms (Table 17).

Figure 28

Table 17: Mean, range, standard deviation, and median first, peak, and last urinary WBC counts in 10 men who underwent a complete remission of symptoms versus 20 men who did not undergo a complete remission of symptoms.

We used the Mann-Whitney test to compare whether the first post-massage urinary WBC counts in the 10 men who reported a complete remission of symptoms (mean, 16.90; median, 6.50) differed from the first post-massage urinary WBC counts in 20 men who did not report a complete remission of symptoms (mean, 7.45; median, 3.00) and the difference was significant (P = 0.032). We compared the peak urine WBC counts in 10 men (mean, 32.00; median, 30.00) to the peak urine WBC counts in 20 men (mean, 30.10; median, 25.00) and the difference was not significant (P = 0.642). We also compared the last urine WBC counts in 10 men (mean, 13.20; median 6.50) versus the last urine WBC counts in 20 men (mean, 3.80; median, 3.00) and the difference did not reach significance (P = 0.184).

We tabulated the post-massage urinary RBC counts for the 30 men (Table 18).

Figure 29

Table 18: Urinary RBCs, first, peak, last, and all data points for the 30 men.

We performed the Friedman test for differences between the first, peak, and last post-massage urinary RBC counts, and a significant difference was present (P < 0.0001) (Table 19).

Figure 30

Table 19: Mean and median, first, peak, and last post-massage urinary RBC counts.

We applied the Wilcoxon signed ranks test with our Bonferroni correction of P less than 0.017 being significant to compare paired groups. The difference between the first post-massage urinary WBC counts (mean, 2.97; median, 2.00) and the peak post-massage urinary RBC counts (mean, 9.70; median, 6.00) was significant (P < 0.0001). The peak urinary RBC counts (mean, 9.70; median, 6.00) and the last urinary RBC counts (mean, 2.03; median, 1.00) were also significantly different (P < 0.0001). The difference between the first urinary RBC counts (mean, 2.97; median, 2.00) and the last (mean, 2.03; 1.00) was not significant (P = 0.069).

We graphed the mean post-massage urinary RBCs at the first, peak, and last visits for the 10 men reporting a complete remission of symptoms versus the 20 men not reporting a complete remission of symptoms (Figure 12).

{image:31}

We also tabulated the mean and median urinary RBCs for the 10 men who reported a complete resolution of symptoms and for the 20 men who did not report a complete resolution of symptoms, and noted that in the group of 10 men the post-massage urine mean and median RBC counts were higher than in the group of 20 men (Table 20).

{image:32}

We used the Mann-Whitney test to look for differences between the RBC counts in the group of 10 men versus the group of 20 men. We compared the first RBC counts in 10 men (mean, 3.80; median, 3.00) to the first RBC counts in 20 men (mean, 2.55; median, 1.50) and the difference was not significant (P = 0.360). We compared the peak RBC counts in 10 men (mean, 12.60; median, 8.00) versus the peak RBC counts in 20 men (mean, 8.25; median, 6.00) and the difference was not significant (P = 0.278). And, we compared the last RBC count in the 10 men (mean, 3.30; median, 1.50) versus the last RBC count in the 20 men (mean, 1.40; median, 1.00), and the difference was not significant (P = 0.469).

Trichomonas was seen in the first post-massage urine of 2 patients (patient Nos. 1 and 27).

Five patients (patient Nos. 1, 10, 11, 25, and 26) had yeast seen in their post-massage urine. Overall, 7 of the 30 men in this study (patient Nos. 1, 2, 10, 11, 25, 26, 30) were reported, by microscopy, to be positive for yeast in their urethral smears, prostatic fluid, or post-massage urine. In the 7 yeast-positive patients, the mean first symptom score was 7.57 (range, 5-10; SD, 2.5), while the mean last symptom score was 3.47 (range, 0-9; SD, 3.47). The decline in symptom scores from first to last was significant (P = 0.033). Four of the 7 yeast-positive patients were in the group of 10 men with complete remission of symptoms.

We list all the medications prescribed to the patients and their frequency of use. All 30 patients were prescribed either doxycycline or minocycline. Twenty-eight of 30 men (93.33%) were prescribed metronidazole. Twenty-eight of 30 men (93.33%) were prescribed antifungals, and many were prescribed both fluconazole and itraconazole. Patients No. 9 and No. 17 were not prescribed antifungals and neither tested positive for yeast during treatment. Cefixime, ofloxacin, and azithromycin were also prescribed frequently. Three patients took the herb, Pygeum africanum, on their own during the study (Table 21).

{image:33}

For additional gross descriptive purposes we again divided the 30 men into two groups: the 10 men reporting a complete resolution of symptoms and the 20 men who did not report a complete resolution of symptoms (Table 22).

{image:34}

We noted that both groups had the same frequency of being positive for WBCs in the urethra at the first visit (30%); however, in the group of 10 men with complete remission, 80% showed urethral WBCs at some point during therapy, while the group of 20 men showed urethral WBCs only 30% of the time.

We tabulated the mean and median WBC and RBC counts for the 10 men reporting a complete remission of symptoms versus the 20 men not reporting a complete remission of symptoms (Table 23).

{image:35}

In the 10 men who underwent complete remission of symptoms, the mean and median EPS WBC counts fell from first to last. This contrasts with the 20 men who did not have complete remission, where the mean and median EPS WBC counts rose from first to last. In almost all cases the first and peak WBC and RBC counts were higher in the 10 men who underwent complete remission of symptoms than in the 20 men who did not report a complete remission of symptoms.

For descriptive purposes, we made a frequency table of all the microbes diagnosed during treatment in 10 men reporting complete resolution of symptoms versus the 20 men not reporting complete resolution of symptoms (Table 24).

{image:36}

In the 10 men reporting complete remission of symptoms, 90% were positive for fluorescing C. trachomatis elementary bodies, and 80% had a negative DFA test before treatment ended (1 patient was not retested at the end of therapy). Of the 20 men without complete remission, only 35% had fluorescing C. trachomatis elementary bodies seen, and only 30% had a negative test result documented before the end of therapy.

The 10 men with complete remission of symptoms were less likely to have Gram-negative bacteria seen in the EPS, and were more likely to have yeast seen in their urethral smears, EPS, or urine than the other 20 men.

For descriptive purposes, we tabulated the antibiotic therapy for the 10 men versus the 20 men (Table 25).

{image:37}

The 10 men reporting complete remission of symptoms received minocycline, cefixime, and azithromycin at greater frequencies than the group of 20 men.

The table also lists the urologic medications given. Only 1 patient in the group of 10 men self-reporting complete remission of symptoms received a urologic medication. Patients who did not undergo complete remission were much more likely to receive alpha-blocker medication or finasteride. This makes sense, because when patients did not improve, the treating physicians often added urological medications.

The 3 patients who took Pygeum africanum on their own did not have an improved remission rate.

Patient No. 1 in this study was the only patient to have a transrectal ultrasound (TRUS) before and after treatment. His presenting prostate size was 41 grams, while his TRUS measured 27 grams after completing therapy, a decline in size of 34.14%. Previously, we published a paper in which the size of a 78-year-old man's prostate declined in size by 52%, from 63 g to 30 g, after repetitive prostate massage and antimicrobial therapy.[₄] Another paper pending publication (Hennenfent BR, Lazarte AR, and Feliciano Jr. AE: Repetitive Prostatic Massage and Drug Therapy as an Alternative to Transurethral Resection of the Prostate) reports decreased TRUS-measured prostate size after repetitive prostatic massage and antibiotic therapy in two men.

Definitions of prostatitis and chronic pelvic pain syndrome are evolving.[₁₀] The National Institutes of Health uses 10 or more WBCs per 400-power magnification as its inflammatory requirement for chronic pelvic pain syndrome; however, the National Institutes of Health criteria do not include repetitive prostatic massage. Schaeffer et al. (Chicago) examined prostatic fluid specimens at high power under a cover slip.[₁₁] The Giessen group (Germany) uses 10 or more leukocytes per 1000-power magnification in the EPS or VB3, after Papanicolaou staining as their definition of inflammatory prostatitis.[₁₂] Krieger et al's work supports the belief that the Gram-stain is a conservative way to count EPS WBCs.[₁₃]

In the current study, 17 of the 30 (56.67%) men had an EPS WBC count of 5 or more based on their first massage. Using all the massages to classify men, 29 of 30 (96.67%) men displayed an EPS WBC count of 5 or more at some point during EPS collection.

Eleven of the 30 (36.67%) men in this study had an EPS WBC count of 10 or more at the first massage, while 26 of the 30 men (86.67%) had an EPS WBC count over 10 at some point during the study.

The 1 patient who did not have 5 or more WBCs in the EPS under 1000X power, patient No. 18, was a 31-year-old man complaining of pelvic pain, irritative urinary symptoms, and obstructive urinary symptoms, whose peak EPS WBC count never rose above 2 WBCs per 1000X, and last EPS WBC was zero. His symptom score dropped from 5 to 2.5 during the course of treatment, but it's doubtful that he had inflammatory prostatitis.

The 10 men who underwent complete remission of symptoms all had an EPS WBC peak over 15 at some point during their EPS collections. In fact, the mean peak EPS WBC count in the 10 men was 42.00 (range, 15–60; SD 18.59; median, 45.00). The mean peak EPS WBC count in the 20 men who did not report complete resolution of symptoms was 33.75 (range, 2–60; SD, 22.10; median, 36.50). However, as noted earlier this difference did not reach significance (P = 0.255).

Because 18 of the 30 patients (60%) were 50 years of age or older, these 18 might have been diagnosed as having BPH by the typical urologist. All 18 patients aged 50 years and older (average age, 61.11 years) had prostatitis by a count of 5 or more WBCs in the EPS, and all 18 such patients also had 10 or more EPS WBCs at some point during therapy. Six of the 18 men aged 50 and older reported complete remission of symptoms; in fact, the average symptom score for these men dropped from 5.78 at the beginning of treatment to 1.51 at the end of treatment (P = 0.00029). The mean peak EPS WBC count in men aged 50 years or older was 41.22 (range 10–60, SD 19.51). We believe that failure to diagnose prostatitis in patients with BPH is a major problem with urological workups, as none of the men over age 50 who presented to the clinic with a diagnosis of BPH had been tested for prostatitis.

The definition and diagnosis of urethritis is also evolving.[₁₄] There is no clear definition of chronic urethritis, as opposed to acute urethritis. None of the patients in this study complained of urethral discharge, while 8 of 30 (26.66%) had a urethral WBC count greater than 0 at their first urethral examination. Three of these men (30%) were in the eventual complete clinical remission group and 5 (25%) were not. Overall, 16 of 30 men (53.33%) were positive for urethral WBCs before treatment ended. This is similar to the National Institutes of Health cohort study, where urethral leukocytosis, defined as 1 or more WBCs in the first voided urine, was found in 243 of 488 cases (49.80%).[₁₅] However, no tests for Chlamydia were done in the NIH study despite the discovered urethral leukocytosis.

In our study, 8 of the 10 men (80%) in the complete remission group had urethral WBCs at some point during treatment. The mean peak EPS WBC count in the 8 men presenting with urethral pus at their first visit was 33.50 (range 5.00-60.00, SD 20.81).

The mean peak urethral WBC count in all 16 men who had urethral pus at some point during treatment was 41.44 (range 5–60, SD 19.93).

It has been hypothesized that repetitive prostatic massage might cause trauma to the prostate. We looked at this by following the RBC counts in the EPS. The data show that while RBCs might be present at the first massage, or during the first several massages, EPS RBCs disappear during therapy. This finding has been consistent in our studies,[₁₆,₁₇] and we believe the improvement in RBC counts demonstrates that prostatic massage is therapeutic rather than traumatic. The occurrence of the WBC peaks and RBC peaks at the same times during the course of massages also serves to indicate that EPS RBCs reflect the inflammation of the prostate.

We believe that repetitive prostatic massage and repetitive EPS collection is critical to the diagnosis and treatment of prostatitis and BPH. In this study, some men might have been classified differently using prostatitis/prostatodynia criteria or the chronic pelvic pain syndrome classification if repetitive prostatic massage had not been performed. Some men would have undergone surgical treatment for BPH instead of medical treatment for chronic prostatitis in others' hands. We believe that our body of work strongly suggests that a one-time examination of the prostatic fluid is insufficient. Only 3 of 30 men (10%) had their highest EPS white blood cell count at their first massage, indicating that several prostatic massages must be done to obtain the most purulent specimen for disease classification and microbial testing.

We were able to obtain EPS in 100 percent of attempts in the 30 men studied, which compares favorably with the National Institutes of Health cohort study, where cohort physicians were only able to obtain EPS in 397 of 488 men (81.35%).[₁₈] The failure of the cohort study to obtain EPS in almost 19% of men (91 of 488 men) puts all their study conclusions in jeopardy, as does basing all their classifications on a 1-time EPS collection. We believe that our ability to obtain EPS in 100 percent of patients over several studies is due to our technique, which is designed to drain the prostate completely of all fluids, and our years of experience doing multiple massages per patient instead of a 1-time massage. A previous study reported that prostatic massage is a poorly studied procedure, and that patients perceive that physicians perform the procedure differently and only sometimes effectively.[₁₉]

Besides obtaining fluid for classification and microbial diagnosis, our WBC count analyses in this paper, and previous papers, and our clinical findings suggest that repetitive prostatic massage unplugs the prostate. This is consistent with work by John W. Polacheck, M.D., and L. Eduardo Vega M.D. (A Beneficial Effect of Prostate Massage for Patients with Chronic Prostatits: The Expression of Intraglandular Prostatic Inflamatory Aggregates, which was presented at the International Prostatitis Collaborative Network Third Annual Meeting, October 23-25, 2000, Arlington, Virginia). Indeed, as we noted earlier, we have documented TRUS-measured decreases in prostate size in some patients after repetitive prostatic massage. In addition, a recent paper demonstrated increased prostatic pressure, presumably from obstruction, and higher WBC counts in men with category IIIa chronic pelvic pain syndrome.[₂₀]

We remind physicians of the age-old surgical caveat, which says that you must drain an infection to cure it. The mean EPS WBC and RBC counts rose and fell during treatment, which we believe suggests that obstructed pockets of purulence are released during the process of repetitive prostatic massage.

One National Institutes of Health cohort study investigator writes: “Although men with CP routinely receive anti-inflammatory and antimicrobial therapy, recent studies suggest that leukocyte and bacterial counts do not correlate with severity of symptoms. These findings suggest that factors other than leucocytes and bacteria contribute to the symptoms associated with CPPS.”[₂₁] However, our study, with statistically significant data, establishes relationships between declining WBC counts and improvements in patient-reported symptom scores. We also discovered that comparing men who reported a complete remission of symptoms versus men who did not report a complete remission of symptoms revealed significant differences in cell counts.

Other researchers have failed to see significant relationships with EPS WBC counts. One study reports: “In cases of abacterial prostatitis, neither the initial expressed prostatic secretion white blood count nor the presence of symptoms reliably predicted subsequent inflammation.”[₂₂] We suggest that repetitive prostatic massage and multiple EPS WBC examinations are needed to see such relationships.

Our study suggests a relationship between fluorescing Chlamydia elementary bodies and symptom improvement, and suggests other possible microbial relationships. In contrast, the National Institutes of Health cohort study failed to test for Chlamydia despite the fact that 49.80% of the men in the study displayed urethral leukocytosis.[₂₃]

By comparing the 10 men who underwent complete remission of symptoms by the end of therapy with the 20 men who did not report a complete remission of symptoms, we saw noteworthy associations. The most successful outcomes were in men whose EPS WBC counts declined the most; who were positive for C. trachomatis by DFA early in treatment; but whose C. trachomatis test turned negative by the end of treatment; and who were treated with metronidazole, cefixime, minocycline, and azithromycin. However, many antibiotics and antifungals were used, and it could have been other combinations that were important.

This study is a retrospective study with all the inherent flaws of such a study. For example, we encountered 12 instances of symptom scores not being documented; however, all the laboratory data on these patients were available. No post-treatment follow-up was performed because of lack of funding. This study also points out the need for a better method to quantify the number of WBCs in specimens when they are too numerous to count under the microscope. A larger study than this one might have found more differences that would have reached statistical significance, and would have allowed more powerful subgroup analyses. More microbial tests, and other tests, could be added to our current protocol if cost were not prohibitive. This study has laid the groundwork for a larger study to be done, and we recommend prospective studies and randomized controlled studies when resources permit.

The authors thank Steve Creech of Statistically Significant Consulting; Hazal Macadangdang, B.S. in statistics, University of the Philippines; Loyd Norella, M.D., MPHM; and Rena Pedaria for their help with statistics. We also thank physicians Bienvenido Garcia, M.D., and Alfred Lazarte, M.D.

This study was not funded, but the authors thank the Prostatitis Foundation (Prostatitis.org), a nonprofit organization devoted to publicizing and researching prostatitis.