Testicular microlithiasis and testicular tumor: a review of the literature

Introduction There are numerous scientific publications on testicular microlithiasis (TML) detected during ultrasound (US) examination. We wished to update the data. Methods PubMed was used to identify original articles published between 1998 and May 2017 describing the association between TML and testicular tumor. Studies were only included if TML was diagnosed by US. Studies were then classified into subgroups according to the following criteria: asymptomatic, symptomatic, infertility, cryptorchidism, family or personal history of testicular cancer, and “no given reason for US”. A Z-Test was used to identify differences within these subgroups. In addition, we identified prospective cohorts of TML patients. Numbers, duration of follow-up, and occurrence of the “testicular tumor” event were recorded for each of them. Results One hundred and seventy-five articles were identified, 40 of which were included. Our review has not showed a clear evidence that cryptorchidism associated with TML is a risk factor for testicular tumor. However, there seems to be a correlation between infertility associated with TML and a higher tumor risk. There were not enough studies to confirm a relationship between family or personal history associated with TML and the tumor risk. There was also a correlation with a higher tumor risk for symptomatic associated with TML and “no given reason for US” plus TML groups. However, these groups are assumed to contain bias and caution must be taken regarding conclusions. Regarding the prospective cohort studies, 16 testicular tumors appeared in the follow-up of patients with TML, 13 patients had risk factors. Conclusion In cases of TML incidental finding by US with the presence of risk factors (personal history of testicular cancer, testicular atrophy, infertility, cryptorchidism) a consultation with a specialist should be considered. In the absence of risk factors, the occurrence of testicular cancer in patients with TML is similar to the risk of the general population.


Background
Testicular microlithiasis (TML) corresponds to concretions of hydroxyapatite surrounded by fibrosis located in the seminiferous tubes [1]. They are due to the insufficient capacity of Sertoli cells to phagocyte the degenerate cells present in these tubes. They are commonly discovered by ultrasound (US). They are not visible on Magnetic Resonance Imaging (MRI). In 1987, Doherty et al. [2] described their appearance on US, which is characterized by a hyperechoic focus measuring between 1 and 3 mm in the testicular parenchyma without posterior shadow cone [3] with a number greater than or equal to 5 per testis. The discovery is mostly fortuitous because there is no clinical manifestation. Their historical radiological classification is described by Backus et al. [4]. Three grades are distinguished according to the number of TML described by parenchyma (grade 1: 5 to 10, grade 2: 10 to 20 and grade 3 with more than 20 TML). In recent years, US has substantially improved with the advent of higher resolutions enhancing TML detection. In 2015, the European Society of Urogenital Radiology (ESUR) proposed a summary of guidelines and reported another classification with 3 groups, based on the number of TML per field of vision [5]. These three groups were defined as follows, limited TML: less than 5 per field of view ( Fig. 1), classic TML: greater than or equal to 5 per field of view (Fig. 2) and finally diffuse TML, labelled "snowstorm" (Fig. 3). There are many observational studies on TML and testicular cancer risk. The objective was to perform a review of the available literature to date.

Search strategy
The literature review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses [PRISMA] guidelines ( Fig. 4) [6]. PubMed was used to identify original articles describing the association between TML and testicular tumor, published between 1998 and May 2017. The following keywords were used in the search strategy: testicular microlithiasis, testicular tumor, testicular cancer, testicular neoplasm. Additional studies were included by analyzing the references cited in the review articles. Relevant studies were selected based on the title and abstract.

Inclusion and exclusion criteria
Studies were included if TML was diagnosed by US. Articles in the English language only were included. Case reports and experimental animal studies were excluded. The following characteristics were collected for each article: year of publication, number of patients included, number of TML carriers, and number of tumor carriers. The studies were classified into subgroups according to the following criteria: asymptomatic, symptomatic, infertility, cryptorchidism, family or personal history of testicular cancer, "no given reason for US". Finally, prospective cohorts of TML patients were also identified. The number of patients, duration of follow-up and finally the occurrence of the "testicular tumor" event were then recorded for each of them.

Statistic analyses
We used a Z-Test to indicate differences in these groups. A P value of 0.05 or less was considered statistically significant. The analysis was performed with Microsoft Excel 2016 (Microsoft, Seattle, WA, USA).

Results
One hundred and seventy-five articles were identified. Of these, 119 were excluded by examining the title and abstract (case reports, written in a language other than English). These articles were then read in their entirety. The studies of Yee et al. [7] and Negri et al. [8] included several risk conditions (infertility and cryptorchidism). The same population was found in two articles [9,10]. A few articles were excluded for missing data.
Finally, a total of 40 articles were selected for our literature review and 135 articles were excluded (Fig. 4).

Asymptomatic cases
Two studies were identified regarding the asymptomatic population [9,11], the TML prevalence was 2.4% [11] and 5.6% [9]. Only one testicular tumor was identified in the TML-free population, and no cases were observed in the population with TML. The pooled data revealed no difference in tumor prevalence within the two groups (NS).

Cryptorchidism
We included 6 studies concerning cryptorchidism [32][33][34][35][36][37]. Two of these series reported a TML frequency of 100% [32,33]. Three cases of testicular tumor only were found in the TML population. No testicular tumor was reported in the TML-free population.  Family or personal history of testicular cancer One study was found regarding TML associated with family or personal history of testicular tumor, the TML prevalence was 48%. Korde et al. [38] reported that TML was more common in the contralateral testis of men with a personal history of testicular tumor. Coffey et al. [39] was not selected because there was no information on whether patients had TML or not. Bach et al. [19] analyzed the association of TML and contralateral tumor in monorchid patients who underwent contralateral orchidectomy for a testicular tumor. Of the 156 patients examined, 23 had TML (15%). A contralateral testicular tumor was diagnosed in 5 patients with TML (21% versus 2% in the TML-free group).

Prospective cohorts of TML
Finally, 16 studies analyzed the occurrence of a "testicular tumor" event in follow-up of patients with TML (

Discussion
In recent years, TML have been the source of several epidemiological studies. Older studies reported low TML prevalence: 1.4% [22] and 0.68% [47]. Prevalence is higher in more recent studies: 12.8% [12] and 18.8% [23]. The advent of new generation probes with improved resolution explains this increase. However, there is a higher prevalence in specific populations at risk: patients with cryptorchidism, infertility, family or personal history, testicular tumor. This raises the question of an association between TML and the risk of developing a testicular tumor. A history of cryptorchidism is a risk factor for testicular cancer [32,48,49]. Negri et al. [8] reported a correlation between germ cell tumor and cryptorchidism associated with TML (odds ratio 7,5 p = 0,04). In our review, there is no clear evidence showing that TML associated with cryptorchidism is a risk factor for testicular tumor. As only a few studies have shown this association, further research should be carried out to confirm it.
Infertility is a risk factor for testicular cancer [50,51]. Some studies have assumed a correlation between testicular cancer and infertility associated with TML [38,52,53]. Our study seems to confirm a correlation between infertility with TML and a higher tumor risk.
Family or personal history is a risk factor for testicular cancer [54,55]. In our review, only one study was identified, however no correlation was found between this factor associated with TML and a higher tumor risk. More studies are required to better assess any potential correlation.
In 2016, the literature review by Pedersen et al. [56] showed similar results. TML are not an independent risk factor for testicular cancer. However, when associated  There are confounding factors regarding the symptomatic group. Some inclusion criteria such as testicular pain, testicular edema or increased testicular volume may reflect the presence of a germ cell tumor and consequently influence the results. These confounding factors are also found in studies in which US is performed without any given indication. Patients included in these cases may have risk factors for testicular tumor.
In a 2015 meta-analysis, Wang et al. [57] concluded that TML have a significant association with testicular cancer. All patients with TML should therefore benefit from close US monitoring. The studies with the most significant forest plot results, Middleton et al. [20], Derogee et al. [21] and Cooper et al. [32], included infertile patients in their samples. The inclusion of studies without distinction of the study population is a confounding factor potentially invalidating the conclusion.
Prospective cohort studies have shown that the occurrence of the testicular tumor event in patients with TML occurred more frequently in patients with testicular cancer risk factors (personal history, infertility, atrophy and cryptorchidism). Patel et al. [58] confirmed the same results in a large retrospective study with a follow-up of 14 years. Among the 442 patients studied, only 2 patients developed a testicular tumor, and both had an independent risk factor of testicular cancer. Furthermore, Pedersen et al. [56] showed that patients often forget to attend their US follow-up. A long term prospective study is difficult to organize.
In 2010, in another meta-analysis, Tan et al. [59] investigated the potential association between TML and intratubular germ cell neoplasia (ITGCN). The study reports a high risk of concomitant discovery of ITGCN and TML when a biopsy is performed on a contralateral testicle of a patient with a history of testicular cancer. ITGCN is where dysplastic cells proliferate inside the seminiferous tubules without crossing the basal membrane. In 2015, Richenberg et al. [5] showed that clustering of TML could cause an unstable area inside the testicle where ITGCN can grow. In patients with a history of orchiectomy for testicular tumors, when TML are present in the contralateral testis, ITGCN is present in 20% of cases. Fifty percent of ITGCN evolve into malignancy within 5 years [60]. A testicular biopsy is then recommended. When an ITGCN is found, therapeutic options can be either external radiotherapy or straight follow up with delayed treatment when a testicular tumor appears. Given the lack of benefit to overall survival, morbidity treatment must be considered, including hypogonadism.
The studies included had different objectives, which may have resulted in selection bias and therefore modify the relationship between TML and testicular cancer. This is the main limitation of the present paper.
We have not studied the histological types of tumor, which may constitute a second bias. Other longitudinal clinical studies should be carried out to determine the association between TML and testicular tumors.

Conclusion
In cases of TML incidental finding by US with the presence of risk factors (personal history of testicular cancer, testicular atrophy, infertility, cryptorchidism) a consultation with a specialist should be considered. In the absence of risk factors, the occurrence of testicular cancer in patients with TML is similar to the risk of the general population.