Successful antibacterial therapy leading ultimately to the cure of CBP has always been hindered by the fact that a large number of commonly used antibiotics, including almost all penicillins and cephalosporins, have proven, with very few exceptions [12] to be ineffective against sensitive uropathogens colonizing prostate glands and ducts. This evidence represents the functional and empirical demonstration of the presence of a “barrier activity” or “fence activity” within the prostate gland, preventing the distribution of blood-borne antibacterial agents to the site of infection. This barrier/fence activity is commonly referred to as the “blood-prostate barrier.”

Barriers aimed at protecting gametes, embryos, and fetuses from external insults are well known and well described in humans as well as in animals. Blood-testis/blood-epididymis and ovarian blood-follicle barriers are typical examples of such structures. However, a comprehensive and systematic investigation aimed at demonstrating the presence and ultimate structure in humans of a blood-prostate barrier at the anatomical and ultrastructural levels has not been performed so far, to the authors’ knowledge.

Nevertheless, there is sufficient evidence at the molecular level to support the hypothesis that a “barrier activity” or “fence activity” may indeed impede the access of a number of antibacterial agents to the lumen of prostate acini or ducts. Such evidence emerged from studies aimed at demonstrating the loss of cell-to-cell connections during the neoplastic transformation of the prostate, enabling cells showing increasingly malignant phenotypes to detach from the primary mass, infiltrate the underlying tissues, and eventually metastasize. In particular, a number of studies focused on the major structural components of tight junctions. Tight junctions (TJs) are closely associated membranes of adjacent cells, joining together to form a barrier which is virtually impermeable to water and many solutes, including most drugs. For example, tight junctions connecting capillary endothelial cells in the central nervous system are the fundamental components of the blood-brain barrier.

Occludins, and the family of claudin proteins, are the major constituents of tight junctions. Bush and coworkers demonstrated that occludin is expressed in normal human prostate glands [7] and shows a characteristic occlusion staining pattern in the apical region of secretory luminal cells. Together with occludin, claudins are the most important components of TJs, and form the paracellular barrier controlling the flux of ions and small molecules through the intercellular spaces between epithelial cells. By immunofluorescence experiments, claudins 1, 3, 4, 5, 7, 8, and 10 were shown to be expressed and arranged in junctional patterns in the mouse prostate epithelium [52], and an effective barrier activity, completely blocking the passage of blood-borne radiolabeled dextran to ductal fluids, was functionally demonstrated in the rat ventral prostate [17]. In humans, claudins 1, 3, 4, 5, and 7 were shown to be expressed and organized in junctional patterns in the apical region of columnar luminal cells in normal, nonneoplastic prostatic epithelia both in prostate cancer patients and in subjects affected by benign prostatic hyperplasia [3, 55, 61] (Fig. 6.2).

The fact that most studies demonstrated junctional patterns of expression of claudins and occludin in the apex region of luminal cells of the prostate glandular epithelium does not restrict the role of “barrier” or “fence” to these cells alone. For example, El-Alfy and coworkers have shown that basal cells within prostate glands, present in an even proportion (~1:1) with luminal cells only in humans, show a strict intercellular continuity ensured by the presence of junction-like complexes, which may indeed further contribute to the “barrier effect” elicited by tight junctions in the apex of luminal cells [14]. In addition, claudin-1 was shown to be preferentially expressed in basal cells in normal human prostates, thus confirming the presence of junctional structures also in deeper, non-apical cellular layers of prostate glands [26].

In summary, molecular evidence supports the hypothesis that adjacent epithelial cells in the prostate gland may be tightly connected to exert a barrier activity, preventing blood-borne drugs to reach the glandular lumen and to target the sites of pathogen colonization and infection. Notably, the presence of tight junction components in the human prostate has been established enough to prompt investigation of claudins 3 and 4 as potential therapeutic targets of cytotoxic C. perfringens enterotoxin, for therapy of primary and metastatic prostate cancer [35]. Moreover, several authors have proposed the use of microbubble-enhanced ultrasound in order to disrupt tight junctions and the “barrier activity” preventing efficient delivery of antibacterials to the lumen of prostate glands. Interesting results have been obtained so far in animal models [34, 54].

To circumvent this barrier activity at the glandular level within the prostate, lipophilic antibacterial agents, able to cross lipid cellular membranes by simple diffusion, must be used in order to achieve drug concentrations sufficient to eradicate the causative pathogens at the site of infection and to prevent the emergence of chemoresistant bacterial strains. In this respect, it was shown that more lipophilic fluoroquinolones like moxifloxacin can achieve higher concentrations in prostatic secretions, compared to less lipophilic molecules like norfloxacin or ciprofloxacin [47, 64].

When chronic infection is caused by intracellular pathogens replicating within luminal and basal cells of the prostate gland (e.g., Chlamydia trachomatis, but also various facultative intracellular bacteria), a therapy with drugs unable to efficiently reach the gland lumen may be attempted, as these agents do not need to cross the junctional barrier located in the apical region of the secretory epithelium to exert their effect. In this respect, macrolides show excellent intracellular penetration and are very effective anti-chlamydial agents in patients affected by chronic prostatitis [36, 58].

Global surveillance studies indicate that fluoroquinolone resistance rates are rising dramatically in almost all bacterial species [13]. In some parts of the world, FQ resistance is increasing in Enterobacteriaceae causing community-acquired or healthcare-associated urinary tract infections, as well as sexually transmitted conditions, and in Asia, resistance levels over 50 % are commonly found.

For example, resistance rates to antibacterials in Neisseria gonorrhoeae can be as high as ~100 %, particularly in Asia; similar to tuberculosis, gonorrhea is nowadays categorized as “multidrug resistant,” “extensively drug resistant,” and “untreatable” [62].

On this basis, guidelines are being adapted worldwide, and alternative protocols are being tentatively suggested, until new studies confirm these indications, or new agents are discovered [41].

In the worrisome era of “superbugs” [9], what are the possible available alternatives to quinolones for managing the diseases treated in this chapter, and in particular CBP?

Indeed, the available armamentarium appears to be quite restricted, and evidence from randomized or cohort-observational studies is lacking. For a limited number of drugs, only published case reports or small patient series are available, and many agents must be administered off-label due to the lack of officially approved and registered indications for prostatitis.

The prostate penetration of meropenem has been documented, and sufficient bactericidal concentrations are achieved in the prostate tissue for most pathogens, albeit the antipseudomonal activity of this drug in the prostate is uncertain [44].

For few aminoglycosides (e.g., netilmicin in some European countries), “prostatitis” is an approved indication, and administration is on-label. Adoption of aminoglycosides for long-term treatment of CBP can be hampered by the risk of toxicity to the kidneys and to the inner ear. Whereas renal damage can be sometimes reversible and may be in part avoidable through careful monitoring and continuous patient hydration, cochleotoxicity and/or vestibulotoxicity can appear abruptly and be severe, leading in some cases to profound, irreversible sensorineural hearing loss [46]. The discovery in the late 1990s of the adenine-guanine substitution at the 1555th nucleotide of mitochondrial DNA as the responsible of the onset of hearing loss, often after the very first dose of aminoglycoside [49], was a major scientific breakthrough, and today pharmacogenetic testing for A1555G, C1494T, and other ototoxicity-predisposing mutations may be performed prior to aminoglycoside therapy [6, 63]. In this regard, the Sanford Guide to Antimicrobial Therapy (Table 10A Antibiotic Dosage and Side-Effects) warns clinicians that one in 500 European patients may have mitochondrial mutations that predict cochlear toxicity [22]. Hence, if pharmacogenetic testing is not feasible, careful exclusion of a history of matrilineal deafness is a strongly recommended assessment prior to aminoglycoside administration.

A drug that may be potentially active in the management of a gram-negative prostatic infection is fosfomycin, administered intravenously or orally; evidence of successful resolution of CBP cases was recently presented, though further investigation is warranted [18, 25].

The emergence of enterococcal and staphylococcal strains with multidrug-resistant ability has greatly complicated the management of prostatic infections, and there is an urgent need to assess the efficacy of alternative agents in the frame of adequately designed, powered, and controlled clinical studies. At present (year 2015), the published evidence is scant, and only case reports and case studies are available. For this reason, any recommendation or suggestion is only tentative and based on limited empirical/anecdotal experience.

Gram-positive pathogens like Enterococcus faecalis, Enterococcus faecium, and Staphylococcus aureus (methicillin resistant) are generally sensitive to agents like linezolid and tigecycline.

The oxazolidinone linezolid is officially indicated for treatment of skin, soft tissue, and lung infections, though its activity against fluoroquinolone-resistant uropathogens has been investigated [45, 65]. Linezolid is bacteriostatic against enterococci and staphylococci, but bactericidal for streptococci. It can be administered orally or parenterally, shows a bioavailability of almost 100 %, and the distribution volume equals approximately the total body water (~40 L). The half-life is about 5–7 h, and twice-daily administration is advised [2, 48]. Metabolic acidosis and myelosuppression are among the more severe adverse effects caused by the drug, though in a limited number of cases. Linezolid, administered intravenously (600 mg, twice daily), in combination with oral co-trimoxazole (960 mg b.i.d.), was effective in resolving a complicated case of prostatitis, likely involving both Enterococci and various Enterobacteriaceae [50].

Tigecycline, a glycylcycline structurally resembling tetracycline, can be used to treat gram-positive, but also gram-negative (including metallo-β-Lactamase multidrug-resistant Enterobacteriaceae, but not P. aeruginosa or Proteus spp.) and anaerobic bacterial infections. It is mainly bacteriostatic, but in few cases it is bactericidal, depending on the targeted organism. Its high volume of distribution of 7–10 L/kg suggests diffuse tissue penetration. Moreover, the drug has a conveniently long half-life (about 40 h). A comprehensive analysis of the potential activity of this agent against CBP, together with a detailed case report, has been published by Bates and coworkers [5]. These authors describe administration of intravenous tigecycline, 50 mg twice daily, after a single “kick-in” 100-mg dose, for 6 weeks.