Outcome of three different techniques of prostatic abscess drainage: a retrospective single-center experience

Background Despite the fact that prostate abscess (PA) is a serious condition, no standardized guidelines are available on the best methods for its drainage. The goal of the current study is to compare the outcomes of the three modalities of prostate abscess (PA) drainage: transrectal ultrasound (TRUS) guided aspiration (TRA), transperineal aspiration (TPA), and transurethral de-roofing (TUD). Methods The records of 66 patients diagnosed with PAs were collected between 2015 and 2022 and divided into 3 groups according to the technique of abscess drainage that was used. Group A included 26 patients treated by TRA; Group B included 28 patients treated by TUD; and Group C included 12 patients treated by TPA. The three groups were compared based on demographic data, drainage success, recurrence, complications, and hospital stay. Results The mean abscess size (cm) was 4.8 ± 1, 5.7 ± 1, and 5.4 ± 1.3 cm in groups A, B, and C, respectively. Central abscesses were detected in 35 patients (53%), peripheral in 21 patients (31.8%), and both central and peripheral in 10 patients (15.2%). The most common clinical presentation was lower urinary tract symptoms (LUTS) in 45 patients (68.18%), followed by fever in 40 patients (60.6%). The mean abscess size (cm) was 4.8 ± 1, 5.7 ± 1, and 5.4 ± 1.3 cm in groups A, B, and C, respectively. Recurrence of the abscess occurred in three (11.5%), one (3.6%), and one (8.3%) patients in groups A, B, and C, respectively ( p = 0.496). The mean hospital stay was 4.8 ± 1.7, 5 ± 1.8, and 4.5 ± 1.3 days in groups A, B, and C, respectively ( p = 0.763). When compared to the other two groups, the TUD group had a higher rate of complications (32.1%). Conclusion A TRUS-guided transrectal and transperineal aspiration of PA were comparable to TUD technique regarding the recurrence rate and hospitalization period; however, the safety profile was in favor of the aspiration groups.

Prostatic imaging studies such as transrectal ultrasound (TRUS), computed tomography (CT), and magnetic resonance imaging (MRI) have become widely used to diagnose PA in the recent years.Particularly, TRUS is among the most frequently used modalities for the diagnosis of PA as it is associated with lower cost, availability, and absence of exposure to radiation.Dell' Atti L [9] proposed a classification of PA into five distinctive categories according to the database of about 1300 TRUS images and based on the size, extension, and focality of the abscess.This uniform classification would make it easier to diagnose and localize PA while guiding physicians to the appropriate therapeutic treatments.
A clinical suspicion of PA should be addressed in highrisk patients who present with lower urinary tract symptoms (LUTS) and are not responding well to medical treatment [2].Early clinical suspicion and diagnosis of PA are paramount to avoid its serious complications including urosepsis and death [10].
There has been no consensus or guidelines for the diagnosis and treatment of PA until now [10].Parenteral or oral antibiotic therapy should be started immediately when there is a clinically suspected PA, followed by drainage of the abscess when the diagnosis is confirmed by imaging, in an attempt to reduce its serious consequences.Transrectal aspiration (TRA), transperineal aspiration (TPA), and transurethral de-roofing (TUD) are the most commonly used techniques for PA drainage.Although TRA and TPA of PA are less invasive therapeutic strategies, they may be associated with a higher risk of incomplete drainage and subsequently abscess recurrence.TUD is more commonly utilized in more complex and multiloculated abscesses [11].Herein, the current study aimed to evaluate the outcomes (recurrence, hospital stay and complications) of three different modalities of PA drainage (TRA vs. TPA vs. TUD).

Type of study and patients' inclusion
After the approval of the local ethical committee (code: 36135/12/22), we retrospectively reviewed the medical records of 73 PA patients presented to our hospital, during the period from December 2015 to December 2022.Seven patients were excluded from the analysis (three patients with small size abscesses responded well to medical treatment with intravenous antibiotics, two patients due to the incompleteness of their medical records, and two patients presented with large extraprostatic abscesses that extended into the perineum and required open drainage).The remaining patients (n = 66) were divided into 3 groups according to the abscess drainage technique used (Fig. 1).Group A included 26 patients that were treated by TRUS-guided TRA; Group B included 28 patients that were treated by TUD; and Group C included 12 patients that were treated by TRUSguided TPA.An informed written consent was taken from all the participants.Also, informed consent for publication was obtained, and the images of participants will be freely available on the internet and may be seen by the general public.

Diagnosis of PA and patients' evaluation
The diagnosis of PA was based on clinical symptoms (LUTS, fever, perineal discomfort, urinary retention, etc.), presence of fluctuation on DRE, trans-abdominal ultrasound, TRUS findings (hypo-echoic lesions with or without thick fluid inside), total leucocytic count (TLC), and urine culture.In cases of unclear diagnosis or suspicion of extra-prostatic involvement, CT and/or MRI of the pelvis were required.Patients' demographics, abscess features, treatment given, abscess recurrence, complications (according to Clavien-Dindo grading system), and hospitalization time, were assessed and compared between the three groups.Success of the procedure was defined as clinical improvement and no recurrence of the abscess on follow-up imaging.
A cardiac consultation was done for all patients with ischemic heart disease who were taking antiplatelets and/ or anticoagulants.Patients who were on low-dose aspirin continued it without significant risk of bleeding.Patients on clopidogrel: we stopped it 5 days before the intervention, and bridging with low molecular weight heparin was done 24 h before the intervention.

Surgical technique
A preoperative antibiotic, 1 gm of meropenem IV infusion, was given to all patients before the drainage of PA.All cases of PAs in this study were treated by the same surgical team.
Group A (TRA): Under local anesthetic, the patient was positioned in the left lateral decubitus.We used the BK Medical Flex Focus 400 ultrasound (Burlington, MA, USA) and its linear transrectal probe (7.5 MHZ).Before the procedure, the patients were encouraged to use an enema to evacuate the rectum, and then we injected 10 mL of povidone-iodine 10% intra-rectal to reduce the risk of infection.Lignocaine gel 2% was injected inside the rectum 5 min before the introduction of the probe to lessen its discomfort.An 18 G Chiba needle (Coloplast, Denmark) was introduced into the abscess cavity via the needle guide affixed to the probe and aspirated manually via a 20-cm syringe fitted to the Chiba needle (Fig. 2).Following that, the abscess cavity was washed with 0.9% saline and 2 mL of gentamicin was injected.The aspirated pus was sent for a culture and sensitivity test.
Group B (TUD): Under spinal anesthesia, the patient was placed in the lithotomy position.The preoperative DRE and TRUS data were used to properly pinpoint the location of the abscess.The intraoperative pus discharge into the prostatic urethra that occurs when prostatic massage is performed may also aid in localization of the abscess.After its identification, prostate tissues around the cavity's mouth were resected using a 26 Fr monopolar resectoscope (KARL STORZ SE & Co. KG-Tuttlingen, Germany) to completely de-roof the cavity (Fig. 3).The end point for resection in TUD was complete de-roofing of the abscess cavity and complete evacuation of the pus helped by prostatic massage.The resected prostatic tissues were sent for histopathology to exclude any suspicion of prostate cancer.A urethral catheter was fixed for all patients.
Group C (TPA): The patient was placed in the lithotomy position, and then the perineal area is prepared with povidone-iodine 10%.For perineal nerve block, 20 mL of 1% xylocaine was injected subcutaneously via a 20 G Chiba needle, and then injected deeply under the guidance of TRUS until the level of the pelvic floor for peri-apical prostatic block.After we properly identified Transrectal probe insertion, after local anesthetic and povidone-iodine application, followed by abscess drainage by manual aspiration via a 20 mL syringe attached to the needle.C TRUS imaging shows complete aspiration of the abscess, and 2 mL of gentamicin was injected the location of the abscess, an 18 G Chiba needle (Coloplast, Denmark) was introduced through the perineal skin into the abscess site under TRUS guidance, and then aspiration was done (Fig. 4).Subsequently, the abscess cavity was washed with 0.9% saline and 2 mL of gentamicin was injected.The drained pus was sent for culture and sensitivity testing in all cases.

Patients' follow-up
Follow-up included: daily clinical evaluation (vital signs measurement), TLC after two days, and TRUS within one week.For patients with a fixed urethral catheter, a trial without catheter was performed five days after the drainage of the abscess; if retention occurred, reinsertion of the urethral catheter was done, and the patient was scheduled for a later transurethral resection of prostate (TURP).The patients continued antibiotics and antipyretics and were then discharged after normalization of TLC and being a febrile for two days, and continued antibiotics therapy according to the culture and sensitivity results for at least two weeks.

Statistical analysis
The IBM SPSS software package, version 20.0, was used to gather, input, and analyze the data (IBM Corp., Armonk, NY).Categorical data were shown as percentages and numbers.The range (minimum and maximum), mean, standard deviation, and median were used to express quantitative data.The Chi-squared test was applied for the comparison of categorical variables.The three study groups were compared using ANOVA.P values below 0.05 were deemed significant.

Results
A total of 66 patients diagnosed with PA who required intervention were included in the analysis.The most frequent clinical presentation was LUTS in 45 patients (68.18%), afterwards fever in 40 patients (60.6%) and perineal discomfort in 38 patients (57.5%).Table 1 summarizes the patients' clinical and demographic data.The mean abscess size (cm) was 4.8 ± 1, 5.7 ± 1, and 5.4 ± 1.3 cm in groups A, B, and C, respectively.Regarding the abscess site, central abscesses were detected in 35 patients (53%), peripheral in 21 patients (31.8%), and both central and peripheral in 10 patients (15.2%).Considering the histopathology of the resected prostatic tissues in TUD group, two patients (7.14%) were diagnosed as adenocarcinoma of the prostate (Fig. 5).The culture of retrieved pus was slightly different from preoperative culture as follows: E. coli in 26 patients (39.3%),Klebsiella in 10 patients (15%), pseudomonas in 7 patients (10.6%),Staphylococcus aureus in 8 patients (12.12%), and no organism (negative culture) in 15 patients (22.7%).Recurrence of the abscess was reported in five patients (7.5%), of which, three in TRA group, and one patient in each of the other two groups (p = 0.496).Four patients who developed PA recurrence were diabetic.The sizes of recurrent abscesses ranged from 2-6 cm. and were peripheral in 4 patients and central in one patient.The timing of recurrence ranged from 1-6 weeks from the time of initial drainage.Transrectal re-aspiration of the recurrent abscesses was done successfully for all cases in groups A and B. The recurrent case in Group C was successfully treated by open perineal drainage because it had an extra-prostatic extension into the perineum, and compressed the rectum posteriorly.In this case, via midline perineal incision, we opened the skin and superficial fascia till the bulbospongiosus muscle, we faced up by the pus discharging from around the urethra, then blunt drainage through the perineal membrane was done to reach the prostatic abscess and drain it.Finally, a drain was inserted (Fig. 6).
The mean duration of hospital stay was 4.8 ± 1.7, 5 ± 1.8, and 4.5 ± 1.3 days in groups A, B, and C, respectively (p = 0.763).No deaths occurred after any of the drainage techniques.Regarding the complications (according to Clavien-Dindo grading system), septic shock (Grade IV a, defined as patients who developed symptoms and signs of sepsis after the drainage and progressed to persistent hypotension despite adequate fluid resuscitation) occurred in one patient in Group A and in two patients in Group B (requiring intensive care unit [ICU] admission).Two patients were diabetic, and one had recent urethral instrumentation.
Urethral stricture (Grade III a) developed in one patient in Group B (treated by visual internal urethrotomy.In Group C, one patient developed a small perineal hematoma (Grade I) that was resolved conservatively with follow-up, and another patient developed erectile dysfunction (Grade II) probably due to nerve damage and treated by phosphodiesterase type 5 inhibitors (Table 2).
We had six cases that required TURP after a failed trial without a catheter.We observed that patients on preoperative urethral catheter (4 patients) were the main risk factor for the later TURP need.Additionally, two patients had neurogenic bladders.

Discussion
Prostatic abscess is relatively uncommon due to the widespread prescription of antibiotics among individuals presenting with LUTS.Despite its seriousness; which could result in sepsis and death; there is no standardized treatment or guidelines for this important clinical entity [12].Several techniques have been described for surgical drainage of PA, all of which have been shown to be effective and practicable.These techniques include TRUS-guided aspiration (either TRA or TPA), TUD, transurethral holmium laser resection, and finally open drainage [13][14][15].
It is crucial to address the risk factors and the associated co-morbidities in patients diagnosed with PA.In our results, diabetes mellitus (DM), chronic kidney disease (CKD), hepatic dysfunction, indwelling urethral catheters, recently performed prostatic biopsies and urethral instrumentation were identified in many patients.Noteworthy, DM is the most commonly associated risk factor for PA, suggesting a potential association between DM and severe infections, such as PA [16,17].
The current study supported this association, where DM was documented in about 50% of the included patients.In line with these results, Walker et al. [18] reported a higher incidence of emphysematous PA in diabetics, with mortality rates as high as 25%.Ha A et al. [19] recently conducted an analysis of the risk factors of PA, and they concluded that history of prostate biopsy, complicated DM, and urethral stricture are the most significant risk factors.There are a number of theories as to why diabetics are more likely to develop PA, including a more favorable microenvironment, a neuropathic bladder that causes urine stasis, and a propensity for skin infections that result in local seeding [17,19].Many publications recommend the TRUS-guided TRA of PA as most urologists are familiar with this technique [20,21].The benefits of this technique include the ability to do aspiration under local anesthesia, enabling the practitioner to diagnose and intervene in a single session, and the ease of repeatability of the procedure in the event of recurrence.Additionally, this less invasive technique is associated with fewer complications, particularly ejaculatory dysfunction in young patients, which is a common complication following TUD of PA [22,23].Abscess recurrence, especially in large and multiloculated abscesses, is still the major drawback of this technique.Recurrence of PA after TRA has been reported to be as high as 15 to 33% in the literature [20,22,24,25].In this setting, Selem et al. [26], Jang et al. [27], and Purkait B et al. [12], reported a significantly higher recurrence rates in patients treated by TRA than in patients treated by TUD.On the contrary, Alnadhari et al. [28] observed no recurrences for patients in both the conservative and TUD groups, and recurrences were noted in only three patients (50%) in the TRUS-guided aspiration group; however, there was no significant difference between the three groups.The current study supported this finding; yet there was a tendency towards higher recurrence rate in the TRA group compared to TUD and TPA (11.5%, 3.6%, and 8.3%, respectively, p = 0.49).Interestingly, abscess recurrence was observed by El-Shal et al. [20] to occur in 7% of patients following TUD without any recurrences following TRUS aspiration, and this was attributed to the existence of numerous abscesses in 50 percent of patients in both groups, which cannot be seen intra-operatively during TUD but can be clearly visible and drained by TRUS.
Despite, TUD of PA may be more suitable for patients with multiloculated abscesses; there is an increased risk of retrograde ejaculation, urethral strictures, and urine incontinence.Patients, especially those seeking conception, may be persuaded to use needle aspiration due to these risks.TUD also needs anesthesia, which can be problematic for patients with hemodynamic instability.It also increases the bleeding risk, especially in those with platelet deficiency, which is frequently associated with sepsis [29,30].As a successful alternative to loop transurethral drainage, transurethral holmium laser deroofing of PA has also been documented in some studies with promising outcomes [15].In the current study, the TUD group was associated with a significantly higher rate of complications (32.1%) when compared to the other two groups; epididymo-orchitis occurred in two patients; septic shock in two patients; hematuria in one patient; urethral stricture in one patient; and two patients reported retrograde ejaculation.In the line of our results, El-Shal et al. [20] described that the complications occurred only in the TUD group: septic shock necessitating ICU admission developed in two patients; epididymo-orchitis occurred in one patient; additionally, one patient experienced urethral stricture necessitating an endoscopic urethrotomy; and one patient experienced urinary incontinence with urethral diverticulum, necessitating diverticulum excision and urethral sling.Lastly, a urethro-rectal fistula occurred in one patient.According to Selem et al. [26], the complications ranged from grade II to IVa on the Clavien grading scale and were more prevalent in the TUD group.
Because of the increasing interest in the transperineal route for prostate biopsy worldwide in an attempt to reduce the infection rate, the TPA of PA has emerged and started to gain some popularity in our center a few years ago.This technique had been discussed in a few studies in the literature and on a small number of patients [14,24,29].Varkarakis J et al. [14] described the use of three dimensional ultrasound-guided transperineal drainage of PA in seven patients with no recurrences or complications.Interestingly, following TPA of PA, Bhagat et al. reported that periurethral or perineal abscesses developed, necessitating an open incision and drainage [29].
The duration of hospital stay is variable in the published studies discussing the different techniques of PA drainage.In this study, the mean hospitalization period was comparable among the three groups: 4.8 ± 1.7, 5 ± 1.8, and 4.5 ± 1.3 days in groups A, B, and C, respectively (P = 0.763).Despite being longer, Alnadhari et al. [28] reported no significant differences between the hospitalization period of different treatment modalities (conservative, TRA, and TUD) with a mean of 8.45 days (range 2-21 days).However, this longer hospitalization period may be explained by the small sample size in each group and that 50% of the patients in the TRA group had recurrent abscess that required TUD, and it may be related also to the hospital protocols.On the contrary, El-Shal et al. [20] reported a significantly shorter hospitalization period in the patients who were managed with TRA as compared to the TUD; this may be attributed to the small number of patients in the aspiration group and the smaller abscess size (2.7 mL) in this group.On the other hand, several authors [12,26,27] reported a shorter hospitalization period in the TUD group compared to TRA group.In general, the fewer encountered complications and recurrences, as well as our center's excellent experience conducting TRA and TPA under local anesthetic, may have contributed to the short hospitalization in our study compared to many previous studies.
To our knowledge, this is the largest and only report comparing the outcomes and complications of three different techniques of PA drainage together so far.However, it is not devoid of limitations, such as; the retrospective design of the study and, the small sample size.Therefore, further prospective large-scale multicenter studies comparing the outcome of the three different techniques are required to confirm the results.

Conclusions
Prostate abscess is an uncommon urological entity with a challenging diagnosis and treatment.TUD, TRA, and TPA are comparable as regards the recurrence rate and the hospitalization period.However, the fewer complications encountered in TRUS-guided aspiration of PA and the feasibility of performing these techniques under local anesthesia may favor these options over TUD.Accordingly, we recommend the TRUS-guided aspiration techniques of PA as the initial procedure for PA drainage, and TUD should be saved for intractable cases.

Fig. 1
Fig. 1 Flowchart and summary of patients' outcome

Fig. 2
Fig. 2 Technique of TRA of PA: A TRUS imaging shows right lobe basal PA at about 4 × 3.5 cm with an 18 G Chiba needle inside (red arrow).B Transrectal probe insertion, after local anesthetic and povidone-iodine application, followed by abscess drainage by manual aspiration via a 20 mL syringe attached to the needle.C TRUS imaging shows complete aspiration of the abscess, and 2 mL of gentamicin was injected

Fig. 3 Fig. 4
Fig. 3 Technique of TUD of PA: A TRUS imaging shows a large, multiloculated PA.B The Cystoscopic view shows a gush of thick pus after resection of the abscess' mouth

Fig. 5 Fig. 6 A
Fig. 5 Histopathology of the resected prostatic tissues in TUD group.A Prostatic acini with hyperplastic epithelial lining (black arrow).B Marked stromal inflammatory infiltrate (black arrow), with abscess formation (red arrow).C Shows well-differentiated prostatic adenocarcinoma with a Gleason score of (3 + 3) in a 64-year-old patient presenting with PA and PSA 16

Table 1
Patients' demographic data and characteristicsTRA transrectal aspiration, TUD transurethral de-roofing, TPA transperineal aspiration, SD standard deviation, AUR acute urinary retention, LUTS lower urinary tract symptoms, DRE digital rectal examination, PSA prostatic specific antigen, TLC total leucocyte count

Table 2
Peri-procedural outcomes and complicationsValues presented as mean ± SD (range) or number (%) TRA transrectal aspiration, TUD transurethral de-roofing, TPA transperineal aspiration, SD standard deviation, TURP transurethral resection of prostate p: p value for comparing between the three studied groups * : Statistically significant at p ≤ 0.05 a Significant with Group A