Testicular microvascular flow is altered in Klinefelter syndrome and predicts circulating testosterone

INTRODUCTION

Cardiovascular disease represents a major factor accounting for increased morbidity and mortality in subjects with Klinefelter syndrome (KS) (1,2). A Danish population study documented an approximately 2-fold increased risk of hospitalization because of circulatory diseases, which were present even before the diagnosis of KS, and a 4-fold increased risk of congenital heart malformations (2). In contrast, a British population study documented a 30% increase in mortality owing to circulatory diseases and a 7-fold increase in mortality owing to cardiovascular congenital anomalies compared with the background population (1). Contributing factors with respect to the heart and vasculature have been confirmed.

In the hearts of subjects with KS, an increased prevalence of mitral valve prolapse has been reported by some earlier studies (3,4), alongside diastolic dysfunction (5,6), chronotropic incompetence (6), and a shortened QTc interval (7). Vascular anomalies intrinsic to KS are evidenced by the finding of reduced luminal diameter of the main arterial blood vessels (namely, brachial artery, common carotid and femoral arteries, and abdominal aorta) (8), which is not explained by testosterone (T) levels or gonadal status and is coherent with a supposed X chromosome gene dosage effect; the opposite is indeed true in Turner syndrome, in which a tendency toward dilation of major arteries is observed (9). Furthermore, venous thrombosis is especially common in KS, with a 4-to 8-fold increased risk, and significantly contributes to the overall morbidity and mortality, independently of testosterone treatment and especially in young men (2,10,11).

Little is known regarding testicular vasculature and vascular abnormalities in subjects with KS and how these factors are possibly related to testicular function and frequent hypoandrogenism observed in this condition (12). Most intriguingly, a preliminary report from 2014 documented higher intratesticular T (ITT) levels in subjects with KS than those in normal controls, despite low serum T levels, resulting in a markedly increased ITT/T ratio (8 to 10 times higher in KS compared to controls), whereas the ITT/luteinizing hormone (LH) ratio was comparable among subjects with KS, patients with Sertoli-cell-only syndrome, and subjects with normal spermatogenesis (13). Under the assumption of normal Leydig cell function, the authors explored a “vascular” explanation for the lack of T release in the bloodstream and found evidence of vascular abnormalities in testis biopsies from adult male 41,XX^Y * vs 40,XY* mice. They observed a decreased blood vessel/testis surface ratio and, thus, speculated whether T production is not actually impaired in men with KS, but its release in the systemic circulation is hampered. They suggest that a reduced vascular bed is responsible for a lower T release into the bloodstream (13). The same group subsequently investigated the timing of vascular alterations during testicular development in mice, and found that such alteration are progressively acquired, with a lower number of small- and mid-sized blood vessels only in adult KS mice. They further conducted a study on testicular blood flow via contrast-enhanced ultrasonography (CEUS) and demonstrated reduced perfusion parameters, in favor of impaired testicular vascularization in KS mice (14).

Conversely, in human testicular biopsies obtained during testicular sperm extraction (TESE) procedure from prepubertal, pubertal, and adult subjects with KS, no differences were observed in blood vessel number or density when compared with age-matched control samples, except an increase in smaller vessels in prepubertal children only (15).

Color-flow systems used for the detection of vascular flow are challenged by the weak backscatter of blood relative to the tissue (−40 to −60 dB). The practical consequence of this challenge is limited sensitivity to low-velocity flow in smaller vessels (diameter < 2 mm), blood flow in deep vessels (diameter > 10 cm), and flow in regions with tissue motion (16). In the last decades, researchers have developed methods to detect perfusion using microbubbles to improve vascular flow on images. Following a bolus injection, the larger feeding vessels of the vascular tree can be visualized until the microcirculation fills. The use of intravascular CEUS has developed considerably and proved to be useful in many diagnostics fields (17-21). CEUS combines US with intravenous administration of ultrasonographic contrast medium comprising extremely small-sized US-detectable microbubbles (<10 μm), thus allowing visualization of the normal parenchymal microcirculation.

Based on these, the present study aimed to assess in vivo testicular blood flow using CEUS examination in adolescent and adult subjects with KS and to correlate vascular kinetics with endocrine function.

MATERIALS AND METHODS

CEUS

US examinations were performed using an iU22 unit (Philips, Bothell, Wash) with compound spatial and pulse inversion imaging and a 7–15-MHz wide-band linear probe. Axial and transverse scans of the testis were obtained based on scanning protocols used by Dogra et al (23) and Oyen (24). Representative images and loops were stored in the picture archiving and communication system for subsequent analysis. Testicular US was performed by two experienced sonographers with expertise in urogenital US and qualified in the use of CEUS (A.M.I. and C.P., both with >15 years of experience). All CEUS scans were obtained using the same machine with a low mechanical index (mechanical index = 0.07). Patients received intravenous injection of a 2.4-mL bolus (5 mg/mL) containing contrast material (SonoVue; Bracco, Milan, Italy), with a very rapid infusion, immediately followed by 10 mL of saline. Perfusion analyses were conducted using the quantification software QLAB (Philips). Curves were obtained by manually placing a region of interest (ROI) to entirely cover the testis, carefully excluding the lesion in the control group. ROIs did not change during the entire clip. Quantitative perfusion parameters were obtained using the gamma variate fit specifically designed for the analysis of perfusion imaging using microbubble agents; the function describes the overall transition of intensity value change during a contrast examination, starting with dark images, shortly followed by a steep increase in the early phase (wash-in) and enhancement with slow decrease over time at the end of the curve. The analyzed kinetic parameters included wash-in time (T_in, in seconds), coefficient of the wash-in slope (Slope_in, in dB/second), peak intensity (PI, in dB), time to peak (TTP, in seconds), mean transit time (MTT, in seconds), time of washout 50% (T_{out 50%}, in seconds), and area under the curve (AUC, in arbitrary units) (Figure 1). The Microvascular Imaging (MVI) plug-in for measuring changes in the image from frame to frame, suppressing background tissue signals, and capturing additional contrast data was employed to dramatically enhance vessel conspicuity and map contrast agent progression.

• Download figure
• Open in new tab

Figure 1. Graphic representation of the time/intensity curve and calculated perfusion parameters:

wash-in time (W-in), the time when the testis first started to enhance, measured in seconds; time to peak (TTP), the time needed to reach the peak intensity, measured in seconds; mean transit time (MTT), the time difference between the time needed to reach the peak intensity and the time of the start of the enhancement for the ROI, measured in seconds; peak intensity (PI), the maximum enhancement of the ROI, measured in decibel (dB); washout time (T-out), the time difference between the respective 50% peak intensity values in the washout and peak intensity value, measured in seconds; area under the curve (AUC), intensities of the whole period of enhancement, measured in arbitrary units; Slope in, the coefficient of the wash-in slope and reflects the mean blood flow velocity in the region of interest, measured in in dB/second.

RESULTS

Age, testicular volumes, endocrine function, and CEUS parameters of patients with KS and age-matched control subjects are presented in Table 1.

As expected, compared with the CNT group, subjects with KS presented significantly reduced testicular volumes (p < 0.001), higher gonadotropin levels (p < 0.001 for both LH and FSH), and reduced serum T levels (p = 0.02) as well as both T/LH and cfT/LH ratios (p < 0.001 for both) but similar SHBG and cfT levels. Specifically, of the 10 enrolled subjects with KS, four were hypoandrogenemic (T levels < 10.4 nmol/L), whereas the remaining six individuals presented T levels within the reference range.

CEUS parameters of testicular parenchymal vascularization

The analysis of CEUS parameters showed slower testicular perfusion kinetics in subjects with KS than in the age-matched CNT group. Specifically, there was a significantly longer wash-in phase of contrast microbubbles and mean testicular transit time, as evidenced by slower T_in (p = 0.008), MTT (p = 0.008), and TTP (p < 0.001) intervals. The washout phase was also significantly slower, with a prolonged T_{out 50%} (p = 0.008). No significant differences were observed with regard to the Slope_in, PI, or AUC. Vascular kinetics are presented in Figure 2, which depicts a representative examination in a patient with KS (upper panels) and CNT subject (lower panels), alongside perfusion curves (subject with KS in red and CNT subject in light blue).

• Download figure
• Open in new tab

Figure 2.

Quantitative contrast-enhanced US of a KS testis (upper panels) and a CNT testis (lower panels). Contrast pulse inversion harmonic time–intensity curves of the KS testis (in red) and the CNT testis (in light blue) show a slower T_in and TTP and longer MTT and T_{out 50%} in the KS testis than in the CNT testis.

MVI depiction of the differences between a KS testis (left) and CNT testis (right) is displayed in Figure 3: at 30 seconds (panel B), the vascular architecture of the CNT testis is already apparent with an organized and regular pattern, whereas the KS testis vessels are only visualized at 45 seconds (panel D), showing an irregular and chaotic vascular pattern.

• Download figure
• Open in new tab

Figure 3.

Microvascular Imaging (MVI) frames of KS testis (left) and CNT testis (right) at 20 (panel A), 30 (panel B), 45 (panel C) and 90 seconds (panel D) after contrast injection clearly show the different behavior of intratesticular vessels. At 30 seconds, the CNT testis is already filled with Sonovue, its vascular architecture is clearly identified, and the progression is organized and regular; conversely, KS testis vessels are visualized at 45 seconds, and at the end of the examination, the map shows irregular and chaotic vascularization.

DISCUSSION

The presence of increased cardiovascular risk in KS is a well-established fact and has been documented in population studies, which have described an increase in morbidity and mortality (1,2). The contributing factors are less known, and although T probably plays a role as a vasodilator, with an apparent protective effect with regard to atherogenesis (25), the aneuploidy condition per se seems to be involved. This consideration occurs owing to the higher incidence of congenital heart malformations (2) and the finding of reduced luminal diameter of the main arteries in subjects with KS, independent of T levels or gonadal status (8). Whether these anomalies extend to testicular vasculature or whether testicular function is affected is a matter of debate.

Although a Doppler US study initially described high-resistance intratesticular blood flow in subjects with KS (26), significant findings were obtained in 2014 from a preliminary report describing surprisingly increased ITT levels in subjects with KS compared with healthy controls (almost 10 times higher) (13). This result prompted authors to search for a possible explanation to this phenomenon, discarding the idea of impaired Leydig cell function in KS. The histological finding of a low blood vessel/testis surface ratio in a mouse model of KS led to the hypothesis that T release, rather than its production, is actually impaired in KS, by means of a slower and reduced testicular blood flow. These results were confirmed in a subsequent study, in which a reduction in small and medium-sized blood vessels in adult 41,XX^Y * mice was described, and the CEUS study of those testes demonstrated reduced perfusion parameters, favoring impaired testicular vascularization in KS (14). However, these results were inconsistent with the findings of another histological study in human KS testicular biopsies from prepubertal, peripubertal, and adult subjects (15).

The present study examined for the first time in vivo testicular vascularization in human subjects with KS by means of CEUS compared with age-matched controls. Subjects with KS were characterized by significantly slower testicular perfusion kinetics, in both the wash-in and washout phases (Figure 2), resulting in prolonged testicular MTT; the testicular Slope_in was not significantly different between the two groups, possibly because of the lack of differences in the number of larger arteries (14), in which microbubbles first appear; the PI and AUC were also not different between the two groups.

Strikingly, faster testicular vascularization times were associated with better endocrine testicular function, especially with serum T levels, which showed a strong relationship with vascular parameters, correlating positively with Slope_in and inversely with T_in, MTT, TTP, and T_out._50% as well as gonadotropins and T/LH and cfT/LH ratios. Using MVI analysis, we observed vessel conspicuity during contrast agent progression and noted the normal vascular architecture of CNT compared with subjects with KS, in whom small vessels were not clearly seen.

To assess the individual contribution of vascular flow on total T levels we first performed a PCA on CEUS parameters in KS and CNT subjects: two PCs were identified, namely PC1, representing time-dependent behaviour and mostly indicative of venous blood flow, and PC2, representing intensity-dependent behaviour, mainly attributable to large arteries blood flow (27). Of note, PC1 was constituted by the same CEUS parameters upon which KS subjects significantly differed from the CNT cohort. We subsequently proceeded to conduct a multiple linear regression analysis to verify the independent role of vascular blood flow on testicular endocrine function, which showed venous blood flow (PC1) and bitesticular volume to be independent predictors of total T levels.

To our knowledge, this is the first demonstration that altered intratesticular vasculature correlates with altered peripheral hormonal release. If the feedback loop is indeed intact, an inefficient hormone release in the circulation could be masked by an increase in LH, stretching the hypothalamic–pituitary–testicular axis to the maximum but still resulting in lower circulating T levels. Therefore, the higher ITT/T ratio observed in KS, according to the data published by Tuttelmann et al. (13), would not be a sign of Leydig cell failure but a sign of failure of the interaction between the T source and distribution system. PCA and linear regression analyses further point towards a role for reduced venous blood flow in regulating T release from the testes.

Our study has some limitations. First, the sample size and age distribution of enrolled men with KS are small, resulting in 17 testes from adolescents and adults; however, this is in line with the preliminary nature of our study. Second, as a control group, we included subjects undergoing CEUS because of nonpalpable lesions of < 0.5-cm diameter which were benign at definitive histological evaluation. Therefore, we do not expect testicular parenchymal vascularization to be affected in these men, although this possibility cannot be completely excluded.

Lastly, kinetic parameters depend on the manufacturer of the US unit, and although our software provides automatic time–intensity curves measurements, it is possible that analyses conducted using different software on other US systems provide different values.

Further studies are warranted to evaluate testicular KS vascularization in the prepubertal and peripubertal timing, establish whether vascular abnormalities are already present at an earlier stage, and determine whether vascular dynamics can be useful in predicting the “testicular catastrophe” occurring in KS and scheduling TESE.

In conclusion, our study provides the first demonstration of impaired testicular vascularization kinetics in men with KS. These alterations are possibly related to T release in the peripheral circulation.