Cross-sectional imaging of complicated urinary infections affecting the lower tract and male genital organs

Non-invasive, inexpensive and repeatable, ultrasound is the mainstay first-line investigation, providing a rapid overview of the entire urinary tract, with successful detection of urinary obstruction and most abnormal collections. The initial diagnostic assessment of C-UTIs affecting the lower urinary and male genital tract should rely on transabdominal urinary tract sonography. Colour Doppler ultrasound (CD-US) with high-resolution linear probes is warranted when symptoms and physical signs suggest involvement of the scrotum and perineum. Furthermore, if allowed by local pain and tenderness, transrectal ultrasound (TRUS) may offer rapid assessment for the presence of abnormal masses or collections in the prostate and seminal vesicles. Finally, sonography often allows for an easier, non-invasive post-treatment follow-up of abnormalities detected at cross-sectional imaging [4, 8, 9].

Multidetector CT reliably provides a panoramic visualisation of the entire abdomen and pelvis, and represents the mainstay modality for the detection of intra-abdominal and urological infections in the majority of patients and situations. CT is superior to ultrasound in its ability to consistently identify calcific stones along the entire urinary system and to identify gaseous collections, and may detect inflammatory changes in the perivisceral fat and abnormal contrast enhancement in inflamed tissues, urothelium and abscess walls [4‐6].

The main indications for obtaining CT scans are suspected or confirmed urosepsis and sepsis from an unknown source: in this setting, CT offers a proven benefit in detecting the infectious focus and the possible underlying structural abnormalities of the urinary tract. In our experience, CT is generally useful, and may thus be recommended in patients with high or persistent suspicion of lower urogenital tract infection despite inconclusive ultrasound findings, and in those patients with suspected or culture-proven C-UTI with ENUC risk factors such as diabetes, immunosuppression, nephropathy, obstructive uropathy, urologic instrumentation and surgery [1‐3].

Furthermore, in our experience, CT signs of clinically unsuspected C-UTIs are often detected incidentally on studies performed to investigate other conditions such as urolithiasis, renal colic, gynaecologic complaints or unspecific abdominal pain. When these abnormalities affect the scrotum, additional CD-US is indicated for diagnostic confirmation [10].

Indwelling bladder catheters should be closed before performing CT, although in most patients with lower urinary tract dysfunction, bladder distension is poorly tolerated. A preliminary unenhanced acquisition is useful for detecting calcific urolithiasis, which may require specific treatment. Unless contraindicated by history of allergy or impaired renal function, administration of 110 to 130 ml of non-ionic low- or iso-osmolar iodinated contrast medium (CM) such as 350 mgI/mL iomeprol, 320 mgI/ml iodixanol, or 370 mgI/mL iopromide is generally beneficial [4, 6]. According to the European Society of Urogenital Radiology (ESUR) guidelines, special care in ensuring adequate hydration before and after CT is recommended to improve urinary tract opacification and to prevent CM nephrotoxicity in septic or dehydrated patients with limited urine output [11, 12].

In patients with clinically suspected or confirmed UTI, a single breath-hold nephrographic CT acquisition encompassing the abdomen, pelvis and perineum is obtained after intravenous CM injection. On our 64-slice CT scanner, the usual acquisition parameters include 80–140 KV according to body size and weight, automated tube current modulation, 0.75 s rotation time, and 64 × 0.625 mm collimation. Additional excretory phase imaging acquired 5–20 min after CM injection visualises the opacified urinary collecting systems, ureter and bladder. In contrast to classical multiphasic CT studies, modern split-bolus CT urography protocols provide combined corticomedullary, nephrographic and excretory imaging with a reduced effective radiation dose in a single volumetric acquisition. Multidetector CT studies tailored to the urinary tract should be routinely reconstructed along axial, coronal and sagittal planes, and may be reviewed interactively at the workstation and complemented by additional focused reconstructions such as coronal maximum-intensity projection (MIP) pyelographic images [13, 14].

CT is very appealing to clinicians and radiologists, since it provides a comprehensive anatomic and functional assessment of the entire urogenital tract, including information on renal perfusion and opacification of the urinary cavities. However, it should be used with caution, given its key intrinsic disadvantage of irradiation risk, particularly in young patients and women of reproductive age. With the above-described protocol, the estimated radiation exposure from each phase of acquisition generally falls in the range of 8 to 12 mGy. Strategies for dose reduction, such as automated tube current modulation or, if available, iterative reconstruction from raw CT data, are recommended [15].

Albeit limited by lengthy examination time and the need for cooperation in acutely ill patients, MRI provides an excellent multiplanar visualisation of pelvic, genital and perineal structures, with superior contrast resolution and better tissue characterisation than CT. However, MRI is insensitive for the presence of gas and calcifications. Unless contraindicated by claustrophobia, early pregnancy, metallic foreign body in vital sites, cardiac pacemaker or other implanted devices, the use of MRI is increasingly appealing to avoid irradiation of the reproductive organs [16].

In the setting presented herein, MRI is recommended as a useful problem-solving modality to investigate disorders of the urinary bladder, prostate and seminal vesicles after inconclusive sonographic and CT findings, or as a first-line examination in younger patients [4‐6].

MRI is arguably the best technique and therefore warranted for the provision of cross-sectional multiplanar evaluation of suspected or known periurethral and penile disorders [8, 17‐19] and abnormalities affecting the perianal and perineal structures [20, 21] and the scrotum [22‐24].

On current high-field-strength MRI scanners, pelvic studies are generally acquired with the use of phased-array coils; the patient lies supine, with a folded towel positioned between the legs to elevate the scrotum, and the penis secured at the midline hypogastrium. Most protocols rely on multiplanar high-resolution T2-weighted sequences, with short-tau inversion recovery (STIR) or spectral fat suppression in at least one plane. Additional static-fluid MR urography sequences may provide a panoramic view of the entire urinary collecting systems, without the use of intravenous contrast. Three-dimensional fat-suppressed gradient-echo T1-weighted sequences such as T1 high-resolution isotropic volume excitation (THRIVE), liver acquisition with volume acquisition (LAVA) or volumetric interpolated breath-hold examination (VIBE) are generally acquired before and after administration of a standard dose of gadolinium-based CM [5, 19, 25, 26].

The majority of patients experiencing lower C-UTIs have a scarcely distensible, thickened urinary bladder secondary to underlying chronic conditions such as detrusor hypertrophy and recurrent infections. Therefore, diffuse mural bladder thickening is commonly observed in patients undergoing CT for investigation of C-UTIs or urosepsis. However, in our experience, significant (over 1 cm) circumferential wall thickening may correspond to acute UTI, particularly when the muscular layer shows poor enhancement from intramural oedema (Figs. 1 and 2). Active infectious cystitis (AIC) is heralded by minimal, uniform thickening and hyperenhancement of the urothelium which lines the inner aspect of the urinary bladder (Figs. 1, 2 and 3). Similar to inflammatory enhancement described in acute pyelo-ureteritis, this appearance may be encountered in CT studies performed for different or unrelated indications, may be confirmed by performing slab maximum-intensity projection (MIP) reconstructions (Fig. 2b), and should be reported as consistent with AIC [4, 28].

In addition, diagnosis of AIC is further suggested by the identification of inflammatory changes as “hazy” increased attenuation of the extraperitoneal perivesical fat planes (Figs. 2 and 4). Particularly frequent in diabetics, emphysematous cystitis is a form of C-UTI in which gas-forming micro-organisms lead to the formation of characteristic air-attenuation linear changes within the bladder wall (Fig. 4) [6].

Compared to CT, MRI (Figs. 3, 5 and 6) allows superior assessment of the urinary bladder wall, which in normal conditions shows uniformly low T2-signal intensity (SI) corresponding to the detrusor muscle. At MRI, AIC is heralded by focal or diffuse mural oedema (Fig. 5) and inflammatory-type T2 hypersignal of the perivesical fat (Fig. 6), which is most appreciable with fat suppression techniques [4‐6].

At times, AIC may be further complicated by the formation of a mural bladder abscess, which characteristically appears as an intramural or exophytic non-enhancing hypoattenuating (10–15 HU) collection with irregular peripheral enhancement, generally developing at the bladder dome (Figs. 2, 3 and 6). The lack of an appreciable communication with the bladder lumen allows differentiation of a mural abscess from an infected diverticulum. A correct diagnosis of bladder abscess is crucial, since long-term catheterisation plus drainage (either percutaneous or surgical) is required to relieve the septic focus [27, 29‐31].

The differential diagnosis of AIC includes bladder carcinoma and certain non-neoplastic disorders which manifest as an abnormal bladder wall with decreased distensibility and focal or diffuse mural thickening [32]. Largely employed as the initial investigation in patients with gross or recurrent microscopic haematuria, multidetector CT depicts transitional cell carcinoma as a uni- or multifocal, generally asymmetric wall thickening, which enhances mostly after a 60-s delay (Fig. 7). Tumour is suggested over infection when a soft-tissue density irregularity is detected at the interface between mural thickening and perivesical fat. Perivesical invasion is obvious when the neoplasia shows overt growth beyond the outer bladder wall contour. MRI is superior to CT in detecting tumour tissue disrupting the hypointense outer muscle layer and extending transmurally into the perivesical fat [33‐35].

Nephrogenic adenoma (NA) and malacoplakia of the urinary bladder occur mostly in diabetic or immunocompromised individuals, and manifest with haematuria, proteinuria and signs of lower urinary dysfunction. The latter is a rare chronic granulomatous condition; the former results from long-term irritation from calculi, infection, injury or previous surgery causing urothelial metaplasia. Both entities show unspecific imaging features such as multifocal asymmetric thickening or intraluminal vegetations, frequently associated with hydronephrosis [32, 36].

Infections such as tuberculosis and schistosomiasis should be suspected in patients from endemic countries: the latter is still prevalent in areas of Africa and the Middle East, and is characterised by a contracted, thick-walled bladder with calcifications, and may be complicated by the development of squamocellular carcinoma (Fig. 8) [37‐39].

Finally, other conditions which can mimic AIC and bladder carcinoma include rare inflammatory diseases such as cystitis cystica, cystitis glandularis, eosinophilic cystitis, chemotherapy (particularly with cyclophosphamide) and irradiation [32, 40, 41].

With the widespread use of antibiotics, C-UTIs involving the prostate have become increasingly rare, and nowadays are primarily caused (almost two-thirds of cases) by E. coli infection. Acute bacterial prostatitis (ABP) is a serious infection which requires intensive parenteral antibiotics [1].

Prostatic abscesses (PAs) may result from unrecognised or inappropriately treated ABP, or may develop shortly after transrectal ultrasonography (TRUS)-guided biopsy. Most cases occur in the fifth and sixth decades of life, and the usual predisposing factors include diabetes, indwelling catheters, urological instrumentation and anatomical abnormalities. The non-specific clinical manifestations largely overlap with those of ABP, and include fever, dysuria or frequency, dull pelvic or perineal pain, rectal tenesmus, prostatic tenderness at digital rectal examination, and sometimes acute urinary retention. Laboratory findings indicate acute UTI, and increased serum prostate-specific antigen (PSA) is common but generally regresses with therapy. Urine cultures are often negative, and may become positive when the abscess opens into the urethra or bladder [42‐44].

PAs may be detected by TRUS as single or multiple hypoechoic areas with thick walls, floating echogenic speckles in the cavity, and poorly defined periphery, with increased colour Doppler signals [45, 46].

Compared to TRUS, CT with multiplanar image reconstruction provides a more comprehensive visualisation of PAs. The usual appearance is single, septated or multiple fluid-like (−19 to 13 HU attenuation) collections, often with perceptible peripheral or septal enhancement, which cause more or less symmetric prostatic enlargement of variable entity (Figs. 4, 9 and 10). In addition, CT may depict signs of extraprostatic penetration (Fig. 10) in the prevesical space, rectum, perineum or ischiorectal fossa, and consistently allows monitoring of changes after medical or surgical treatment (Figs. 4 and 10). The differential diagnosis should consider prostate carcinoma, particularly with regressive changes after treatment (Fig. 11) [47‐50].

Misdiagnosis or incomplete treatment of prostatic infections may result in serious morbidity, and the differentiation between ABP and PA is crucial, since it impacts management. Parenteral antibiotics plus transrectal ultrasound-guided or transperineal CT-guided drainage of abscesses are generally recommended as the initial approach in order to prevent sepsis. Alternatively, transurethral incision, deroofing or formal surgical evacuation may be required [1, 45, 46].

In the current antibiotic era, seminal vesicle abscesses (SVAs) are even more uncommon than PAs, and are frequently associated with ABP or epididymo-orchitis. SVAs may present with haemospermia or unspecific symptoms of UTI. However, both PAs and SVAs are often insidious and clinically unsuspected; therefore, a high index of suspicion is recommended when interpreting CT studies with the usual risk factors [42‐44, 51‐53].

Transabdominal or transrectal ultrasound may detect SVAs (Figs. 12 and 13) as hypo-anechoic masses [53, 55]. At cross-sectional imaging, the normal seminal vesicles appear as paired, finely septated fluid-containing structures, measuring approximately 3×1.5 cm in adults, and may be moderately asymmetric in size and tend to shrink with advancing age. Their content is normally homogeneous, with fluid-like CT attenuation and low T1- and fluid-like T2-hyperintense MRI signal with enhancing walls and septa after intravenous contrast [5, 52].

Cross-sectional CT imaging depicts SVAs as uni- or bilateral gland enlargement with thick irregular enhancing wall, internal hypoattenuating regions and adjacent fat inflammatory changes (Figs. 12 and 13). Imaging features of bladder and prostatic infection are frequently associated [52, 54, 56].

The reported MRI findings of subacute and chronic seminal vesicle infections include fluid-like septated collections, with more or less thick and marked peripheral enhancement [57, 58]. The differential diagnosis of seminal vesicle infections encompasses congenital cysts, acquired cystic dilatation, chronic tuberculosis, benign (cystadenoma, leiomyoma, teratoma) and malignant tumours, and metastases from prostate, bladder or rectal cancers [5, 52].

Although a conservative therapeutic approach is increasingly adopted, SVAs often require percutaneous transvesical drainage, transrectal aspiration or surgical incision [43, 44, 59].

Spread through sexual contact, urethritis from Neisseria gonorrhoeae and Chlamydia trachomatis is favoured by promiscuity and low socioeconomic status. Alternatively, urethral infection increasingly results from intermittent or permanent catheterisation or from urologic instrumentation. Symptoms include mucopurulent discharge, alguria, dysuria and urethral pruritus [1, 7].

Acute urethritis is generally diagnosed on the basis of clinical and laboratory findings, but imaging may be required to exclude complications. In the past, conventional CM-enhanced radiographic studies were the primary modalities for imaging of the male urethra, particularly for assessment of traumatic injuries, strictures, and abscesses draining into the urethra. However, retrograde urethrography and voiding cystourethrography were unable to assess the periurethral structures [17, 60, 61].

Ultrasound of the penis may demonstrate a periurethral abscess, but is generally cumbersome because of inflammatory swelling and tenderness of the penile and perineal structures. The use of MRI can effectively visualise abnormalities of the periurethral structures [19]. In our experience, acute urethritis appears as diffuse thickening of the urethra and periurethral tissues, with intermediate to high signal intensity on T2-weighted images and intense contrast enhancement (Fig. 14) [5, 62].

Urethritis may be further complicated by a periurethral abscess through infection of Littré's glands. Since the penile tunica albuginea prevents the dorsal spread of infection, abscesses tends to track ventrally along the corpus spongiosum. When Buck's fascia is perforated, the process leads to fasciitis and gangrenous necrosis of the subcutaneous tissue [1].

MRI visualises perineal and penile abscesses as fluid- or pus-filled cavities with enhancing periphery, in communication with the urethra (Fig. 15), and may clearly reveal involvement of the corpora cavernosa and fibrous tunicae. A urethral diverticulum, most commonly located in the distal urethra, may mimic the appearance of an abscess. Treatment requires antibiotics, suprapubic urinary drainage and surgical evacuation [5, 24, 62, 63].

Colour Doppler ultrasound (CD-US) remains the primary modality for investigating abnormalities of the testis and epididymis, particularly to differentiate torsion from inflammatory conditions [4, 8, 9].

Acute epididymitis is almost always unilateral and has a bimodal distribution, with the majority of cases occurring between 16–30 and 51–70 years of age. Whereas in the former age group, infection is commonly caused by Chlamydia trachomatis or Neisseria gonorrhoeae and is associated with sexual activity, in individuals at an advanced age or immunosuppressed individuals, it results from C-UTI by common aerobic urinary pathogens such as E. coli [1, 64].

Clinical manifestations of epididymo-orchitis include scrotal pain, sometimes radiating to the groin or lower abdomen, fever and other symptoms of UTI, variable scrotal swelling and tenderness at physical examination. Laboratory tests, urethral swab and cultures help to direct antibiotic therapy, which cures infection, relieves symptoms and prevents complications and transmission [1, 64].

When CD-US is not the initial examination, careful scrutiny of CT studies may reveal spermatic cord asymmetry with variable thickening and increased enhancement of vessels on the affected side: this rather subtle finding reflects infectious hyperemia and is strongly associated with ipsilateral infectious funiculitis, epididymitis and/or orchitis; hypervascularity of the epididymis may also be observed at CT (Figs. 16 and 17). Alternatively, spermatic vascular engorgement may reflect the presence of a testicular tumour [10, 65].

Similarly, MRI may depict an enlarged epididymis with increased or heterogeneous T2 signal intensity and engorged vessels [5, 22, 24]. The diagnosis of epididymitis is easily confirmed by CD-US, with the usual features including segmental or global enlargement, inhomogeneously hypoechoic and hypervascularised compared to the testis, commonly associated with thickening of scrotal tunicae or hydrocele (Figs. 16, 17 and 18) [8, 9, 61].

Contiguous infectious involvement of the testis (orchitis) is rather uncommon compared to epididymitis, and generally has similar or more severe clinical and laboratory manifestations [1, 64].

CD-US hypervascularity is the key sonographic finding which allows differentiation of acute orchitis from torsion and infarction [8, 9].

At CT, the normal testes are symmetric and hypoattenuating, and are poorly differentiated by thickened scrotal tunicae and hydrocele. On cross-sectional imaging studies, acute orchitis is suggested by asymmetry with enlargement and increased contrast enhancement of the affected testis compared to the contralateral structure; signs of funiculitis and epididymitis are generally associated (Figs. 17 and 18). In normal conditions, at MRI the testes appear homogeneous, T1-isointense to muscle and T2-hyperintense, and the albuginea and mediastinum testis are identifiable as low-signal bands. Testicular inflammation is better demonstrated by MRI as decreased T1 and increased T2 signal intensity compared with the normal testis, with either intense homogeneous enhancement or the characteristic “tiger skin” post-contrast pattern corresponding to preserved septa. Focal or diffuse orchitis may be challenging to differentiate from testicular tumours, which generally show a mass effect and solid-type CT attenuation and MRI signal features, and are not associated with clinical and biochemical signs of infection [5, 22‐24].

Untreated epididymo-orchitis may be further complicated by testicular necrosis and/or development of a scrotal abscess or pyocele, which requires surgical treatment [1, 64]. Abscesses may be sonographically appreciated as an ill-defined lesion with low echogenicity and absent internal flow signals [4, 8, 9]. CT and MRI features of pyocele include complex, heterogeneous fluid collections, surrounded by an enhancing periphery (Fig. 18) or by hyperaemic inflamed surrounding parenchyma [22‐24].

Fournier’s gangrene (FG) is a rare, life-threatening necrotizing polymicrobial infection of the perineal, perianal and genital structure, commonly occurring in diabetics, with a striking male predominance (male-to-female ratio, 10:1). FG may further complicate a lower UTI, or originate from different infections such as colorectal infections, anal fistulas or pressure ulceration. Clinical presentation includes local pain, swollen oedematous or gangrenous overlying skin, sometimes with appreciable crepitus, and progressive hyperthermia. FG represents a surgical emergency: aggressive surgical debridement and broad-spectrum antibiotics are required to prevent a fatal outcome [66, 67].

CT is by far the best modality for imaging of FG; it quickly and comprehensively visualises the extent of FG involvement. CT features include subcutaneous fat stranding at the involved areas, fascial thickening, superficial or deep fluid and air-attenuation collections. Subcutaneous emphysema produced by anaerobic bacteria is the hallmark of FG and is present in approximately 90 % of cases (Fig. 19). Furthermore, CT can define the starting point of the infectious process, thereby allowing differentiation of complicated perineal infections from urinary versus an alternative source, particularly cryptogenic perianal sepsis (Fig. 20) [66‐68].

Another uncommon differential diagnosis of perineal and scrotal infections is hidradenitis suppurativa (HS), an inflammatory disease of the skin and subcutaneous tissues with unclear pathogenesis and chronic progressive course. Mostly encountered in males and black people with poor hygiene, HS presents with tender subcutaneous nodules, which progress to form painful, deep dermal abscesses, sinus tracts, and eventually ulceration and fibrosis. Recurrence is common despite surgical excision [69, 70].

In our experience, MRI may accurately describe the affected regions, and confirm HS over orchiepididymitis and scrotal abscess by demonstrating that tissue inflammation and abscesses are confined to the superficial planes, with a characteristic symmetric distribution and lacking communication with pelvic organs (Fig. 21) [55, 71‐73].