Cystic lesions in the male pelvis have been considered a relatively rare disorder. They are classified according to their location as midline or off-midline, their position relative to the prostate gland, and their embryologic origin [1]. Moore [2] reported an approximately 1% incidence of congenital prostatic cysts at autopsy. It has been reported that medial prostatic cysts may be observed in 5% of outpatients with urologic symptoms [3]. The incidence in apparently healthy men is, however, unknown.

During routine health checkups, we examined men with transabdominal sonography to determine the incidence of midline prostatic cysts in apparently healthy men and to investigate the size and characteristics of these cysts. We also observed changes in cyst size during follow-up examinations.

We designed a prospective study of the incidence of cysts in which the sonographer specifically examined each patient for the presence of midline prostatic cysts. The study included a total of 1,826 transabdominal sonographic examinations performed on 1,115 men who visited the health care department of our hospital for routine health checkups during a 3-year period (from April 1999 to March 2002). A transabdominal sonographic examination is a routine component of most health checkups in Japan for patients older than 35 years. Since May 1995, in addition to upper abdominal mass screening, our hospital has included lower abdominal scans in routine health checkups. All subjects provided written informed consent for participation in this study. Because this was a study of healthy patients during a routine health visit, we chose the noninvasive transabdominal sonography instead of transrectal sonography. Patient age on the first visit for the health checkup ranged from 35 to 65 years (mean, 51 years). The follow-up examination period ranged from 7 to 26 months (mean, 16 months).

In this study, midline prostatic cysts were defined as hypoechoic to anechoic cystic lesions located in the midline of the prostate detected during transabdominal sonographic examination. Midline prostatic cysts were diagnosed on the basis of transverse scans showing cystic lesions in the midline of the prostate gland and longitudinal scans revealing cystic lesions within the prostate gland or between the prostate gland and the seminal vesicles (Fig. 1A, 1B). During the examination, the sonographer decided whether a cyst was present. A patient was recorded as having a midline prostatic cyst when the cyst was visible in both transverse and longitudinal scans. In those cases, the two scans were then printed out as records, and the attending physician made the final diagnosis. In all cases, the physician's diagnosis of the presence of a midline prostatic cyst confirmed the initial observation by the sonographer. In some cases, a cystlike object was detected but was only visible on either the transverse or the longitudinal scan. The image was added to the subject's record, but these patients were not recorded as having a midline prostatic cyst.

Detection of midline prostatic cysts is limited by the resolution of sonographic systems. In our experience, cysts smaller than 3 mm in diameter are difficult to detect, and the internal echo characteristics of such small cysts are difficult to define. Therefore, we have included only cysts with a diameter equal to or greater than 3 mm in this study (Fig. 2A, 2B). The image quality of each detected cyst was assessed on the basis of an objective scale ranging from excellent to suboptimal. We then investigated the midline prostatic cysts for age distribution, incidence, cyst size and shape, posterior echo enhancement, and prostate weight. In addition, changes in cyst size and shape were also observed in 30 cases in which the original patients returned for follow-up examinations.

All transabdominal sonographic examinations were performed with the full-bladder technique. When the bladder was not full, the subject was asked to drink as much water as possible and wait for reexamination. Sonographic images were obtained with three scanners: U-sonic RT 4600 (General Electric Medical Systems, Milwaukee, WI), LOGIQ 7 (General Electric Medical Systems), and HDI 5000 (ATL, Bothell, WA). The scanners were equipped with convex probes of 3.5, 3.5, and 5-2–MHz, respectively.

Transabdominal sonography revealed midline prostatic cysts in 85 (7.6%) of 1,115 men. Table 1 shows the age distribution and incidence of midline prostatic cysts. The age of the subjects with midline prostatic cysts ranged from 35 to 65 years (mean ± SD, 51 ± 7 years). Prostate weight determination was based on transabdominal sonographic measurements and calculated by using the formula for a prolate ellipse:where d₁ is the transversal; d₂, the anteroposterior; and d₃, the sagittal diameter [4]. The prostate weights of subjects with midline prostatic cysts ranged from 11 to 46 g (mean weight, 21 ± 6 g). In Japan, prostate gland size of approximately less than 20 g is considered within the limit for a healthy prostate [5]; thus, our first weight group ranged from 11 to 20 g. The incidence in prostate gland weight less than or equal to 20 g was 5.4% (47/871) and in prostate size from 21 to 58 g was 15.6% (38/244) (chi-square test, p < 0.01).

The midline prostatic cysts detected on transabdominal sonography were all monolocular and had smooth borders. Morphologically, the cysts were round on transverse scans and oval or beaklike on longitudinal scans. The cyst size ranged from a minimum of 3 × 3 × 4 mm to the maximum of 12 × 13 × 17 mm, with a mean diameter of 7 mm transversely, 7 mm anteriorly, and 9 mm sagittally. Forty-two cysts were clearly shown with excellent image quality, 41 were fair, and two were suboptimal. Internally, the 85 detected cysts ranged from hypoechoic to anechoic, of which 56 showed posterior echo enhancement.

The cysts were either completely intraprostatic in 81 cases (95.3%) or partially exophytic between the prostate and the seminal vesicles in four cases (4.7%). We did not observe any exophytic cysts above the prostate. Although we did not include it in this study, we did find one case of a cyst in the off-midline position. In addition, there were no off-midline cysts among those subjects with a midline prostatic cyst.

In 28 (93.3%) of 30 cases, no change in cyst size was observed between the original presentation and the follow-up examinations. However, two cases (6.7%) did show a small increase in size. Figure 3A, 3B, 3C, 3D shows cyst increase during a 1-year period; Figure 4A, 4B, 4C, 4D shows cyst increase over an 8-month period.

Midline prostatic cysts are more common than previously described and are found in 7.6% of asymptomatic men. The first major study of the incidence of congenital prostatic cysts was performed by Moore in 1937 [2]; Moore found cysts at autopsy in seven of 678 cases. The incidence of approximately 1% continues to be the standard quoted figure. In 1985 in Japan, Kitahara et al. [6] performed endorectal sonography on 660 patients who complained of dysuria, and they found cystic lesions on the midline in the posterior upper region of the prostate in seven patients (1.1%). In 1990, Higashi et al. [7] performed transabdominal sonography on 624 patients referred to the department of radiology and found midline cystic lesions in seven patients (1.0%), which is consistent with the reported values of 1% by Slocum in 1954 [8]. However, in 1996 Dik et al. [3] examined 704 patients with symptoms of bladder outlet obstruction or lower urinary tract symptoms using transrectal sonography and reported 34 (5%) with a medial prostatic cyst. This incidence of 5% was found in a study of symptomatic patients; however, to our knowledge, the incidence in healthy men has not been previously reported.

In our study using transabdominal sonography on 1,115 asymptomatic men, we found 85 cases (7.6%) of midline prostatic cysts. Compared with the size of midline prostatic cysts (15–30 mm in diameter) in the study by Kitahara et al. [6] of a symptomatic population, the cysts observed in this study were smaller (from 3 × 3 × 4 mm to 12 × 13 × 17 mm; mean, 7 × 7 × 9 mm). The difference may be associated with whether clinical symptoms exist. Regarding the relationship between location and size, the cysts found between the prostate gland and seminal vesicle (mean, 10 × 10 × 13 mm) were larger than the intraprostatic cysts (mean, 7 × 6 × 9 mm). In addition, two patients in this study showed a small increase in cyst size over a period of time (Figs. 3A, 3B, 3C, 3D and 4A, 4B, 4C, 4D). From the observed growth in one patient (Fig. 4A, 4B, 4C, 4D), we speculate that midline prostatic cysts might grow toward the seminal vesicle.

We found that the incidence of midline prostatic cysts showed a bimodal distribution across age groups. The incidence in the youngest age group of 35–40 years was 10.1%, and the incidence in the oldest age group of 61–65 years was 11.6% (Table 1). In the case of prostate size and rate of detection, we found a statistically significant increasing trend in the incidence of cysts with increasing prostatic weights. We consider as a factor contributing to our finding of an overall higher incidence of detection (7.6%) the possibility that more cysts are found in recent studies because of the improved quality of sonographic systems and the use of higher frequency wideband probes.

Some cystic lesions in the male pelvis may be discovered by chance with sonography that is now more widely used; in other cases, however, they may remain overlooked. Although transabdominal sonography is a common screening technique, it is difficult to obtain clear images of the pelvic area or detect cystic lesions when the bladder is not full. To improve the detection ratio of midline prostatic cysts, radiologists need to be fully aware of the presence of asymptomatic cysts and use transabdominal sonography with the full-bladder technique.

A limitation of this study is the lack of pathologic correlation with the sonographic findings. The actual cause of our cystic lesions could only be defined with more invasive procedures such as urethrography, seminal vesiculography, or cyst puncture. However, given the asymptomatic nature of the patients, such correlation could not be obtained.

In conclusion, midline prostatic cysts represent a more common variant than has been previously reported. The midline prostatic cyst has rarely been found, perhaps because of its smaller size and absence of noticeable symptoms in patients. In our series, cysts were present in 7.6% of asymptomatic men. Thus, in patients with urologic symptoms, detection of midline cysts will require a focused examination to determine whether the cysts represent a normal variant or are the cause of the symptoms. Future studies are needed to determine which sonographic characteristics may be useful to distinguish the normal variant cyst from the cyst that is responsible for urologic symptoms.

We thank Shogo Shimamura for valuable advice throughout this investigation.

Address correspondence to M. Ishikawa.