Introduction

Stroke is a disease with heterogeneous pathogenesis. And it always presents protean clinical manifestation and sometimes confuses with stroke mimics [1]. These are likely to cause a difficulty in identifying the etiology of stroke accurately and rapidly, especially in the hyperacute stroke setting. As we know, appropriate and effective treatments of stroke are adopted according to its etiological subtypes, which will do much to its prognosis and outcome [2]. So accurate and rapid assays to identify different types of stoke is very important. Previous studies have evaluated brain imaging or some biomarkers to differentiate stroke and its subtypes. Nevertheless some of these methods for identifying different stroke subtypes have limitations. Specifically, they may be not valuable and sensitive in the early stages or in mild stroke or in patients with contraindication [3-5]. So it is of great significance to seek more sensitive, comprehensive and cost-effective methods for identifying stroke subtypes.

Corin, a type transmembrane serine protease that primarily expressed in cardiomyocytes, has been identified as physiological natriuretic peptides convertase [6,7]. To literature, corin levels were decreased in some cardiovascular diseases, e.g. heart failure, acute coronary syndrome [8,9]. We previously found that the level of serum soluble corin was also decreased in patients with stroke [10]. In ischemic stroke subtypes, cardiogenic sources account for about 10-26% among Chinese populations [11]. In other words, corin may be closely associated with embolic subtype of ischemic stroke. Consequently, we hypothesized that serum soluble corin level may be varied among stroke subtypes and perhaps used as a differentiator in stroke. However, serum soluble corin levels in stroke subtypes has not yet been studied in humans. Therefore, we studied serum soluble corin levels in 4 stroke subtypes: hemorrhagic, thrombotic, embolic and lacunar stroke.

Methods

Study population

The study was evaluated and approved by the Ethics Committee of Soochow University. And all subjects provided informed consent. This study consecutively recruited 597 patients with first-ever ischemic stroke (481) or hemorrhagic stroke (116) onset within 48 hours confirmed from 3 hospitals from January to May 2014. The inclusion criteria were as following: (1) Age ≥22 years; (2) Stroke onset within 48 hours confirmed by imaging; (3) Able and willing to sign informed consent by patients or their direct family members. Patients with one of the followings were excluded: (1) Recurrent stroke; (2) Current pregnant women; (3) Unable to participate in the follow-up examination. Stroke was diagnosed by an trained neurologist and confirmed by brain computed tomography (CT) or magnetic resonance imaging (MRI) scan.

Definition of subtypes of stroke

The patients were divided into hemorrhagic and ischemic stroke. And ischemic stroke was further divided into 4 subtypes —thrombotic, embolic, lacunar stroke and undetermined subtype according to the trial of ORG 10172 in acute stroke treatment criteria (TOAST) [12] and Cerebral Embolism Task Force for cerebral infarction subtypes [13].

Thrombotic stroke: Thrombotic stroke patients have clinical and imaging evidence of significant stenosis or occlusion (more than 50%) of a major brain artery or branch cortical artery. The clinical syndromes include brain stem or cerebellar dysfunction or cerebral cortical impairment, such as neglect, aphasia. Brain stem or subcortical hemispheric infarcts >1.5cm in diameter on CT/MRI are regarded as potential large-artery atherosclerotic origin. The supportive evidence include previous transient ischemic attack in the same vascular territory, a carotid bruit etc. Potential sources of cardiogenic embolism should be excluded.

Embolic stroke: Brain imaging and clinical syndromes are similar to thrombotic stroke. And patients should be identified with at least one sources of cardiogenic embolism. According to the report of the Cerebral Embolism Task Force [14], the patients with more than 2 primary clinical features or 1 primary clinical features and ≥2 secondary features can be diagnosed as this category. The primary features include abrupt onset of the maximal neurological deficit, presence of cardiogenic embolism and multiple brain infarcts. The secondary features involve hemorrhagic infarct, absence of atherosclerotic vascular disease, cardiac thrombi showed by echocardiography, CT or MRI etc., evidence of vanishing occlusions and embolism to other organs.

Lacunar stroke: Brain imaging shows a relevant brain stem or subcortical hemispheric lesion (<1.5 cm in greatest diameter) or a normal image. The patients have one of the traditional clinical lacunar syndromes (ataxic hemiparesis, pure sensory stroke, pure motor hemiparesis, etc.) without the evidence of cerebral cortical dysfunction.Brain imaging shows a relevant brain stem or subcortical hemispheric lesion (<1.5 cm in greatest diameter) or a normal image. The patients have one of the traditional clinical lacunar syndromes (ataxic hemiparesis, pure sensory stroke, pure motor hemiparesis, etc.) without the evidence of cerebral cortical dysfunction.

The patients with two or more potential causes of stroke or the cause cannot be diagnosed should be classified as undetermined subtype. There were 11 (2.3%) ischemic stroke patients with undetermined etiology, therefore they were excluded.

Data collection

Data on clinical characteristics were recorded in accordance with previous published reports [10]. Blood samples were drawn immediately after hospitalization before receiving any drugs. And serum soluble corin measurements were blinded from the trained neurologist. To report, soluble corin was stable in blood samples frozen at -80°C after several cycles of freezing and thawing [15]. We used a quantikine human corin ELISA-based assays (R&D Systems, Inc., Minneapolis, USA. Catalog: DCRN00) to test soluble corin in serum. All the samples were processed in a duplicate assay. A standard curve was constructed and from which corin concentrations of unknown samples were determined. Intra- and inter-assay coefficients of variation were less than 2.7% and 6.3%, respectively.

Statistical analysis

Data were analyzed using SAS statistical software (version 9.1, Cary, North Carolina). Continuous data are presented as mean±SD (normally distributed data) or median (interquartile range, non-normally distributed data). Categorical data was presented as counts (%). Baseline characteristics between hemorrhagic stroke patients and ischemic stroke patients were analyzed by t-test for continuous variables with normal distribution, the Wilcoxon rank-sum test for continuous variables with skewed distribution, and the chi-square test for categorical variables. Log-transformed serum soluble corin in different stroke subtypes were analyzed in men and women, respectively. T-test and covariance analysis were used to compare log-transformed corin levels of thrombotic, embolic and lacunar stroke patients with hemorrhagic stroke patients, respectively. In ischemic stroke subtypes, one-Way ANOVA test and covariance analysis was used. The adjusted factors included age, body mass index, systolic blood pressure, triglycerides and fasting plasma glucose. All probabilities were two-tailed and P values0.05 was statistical significantly.

Results

Baseline characteristics

We studied 116 (19.8%) hemorrhagic stroke patients and 470 (80.2%) ischemic stroke patients. Baseline characteristics for these participants are listed in Table 1. There were no significant difference between the two groups in gender, cigarette smoking, hypertension, body mass index, waist circumference, total cholesterol and low density lipoprotein cholesterol. Systolic blood pressure, diastolic blood pressure, high density lipoprotein cholesterol, fasting plasma glucose, national institutes of health stroke scale score were all significantly higher while triglyceride was significantly lower in hemorrhagic stroke patients than ischemic stroke patients (all P <0.05). Patients with ischemic stroke were more likely to be older, have a history of coronary heart disease, family history of stroke and diabetes compared with hemorrhagic stroke patients (all P <0.05). Alcohol consumption was more often noted in hemorrhagic stroke patients (P <0.05).

Distribution of serum soluble corin by genders

Table 2 shows the serum soluble corin levels in ischemic and hemorrhagic stroke patients among men and women, respectively. Among men, the median level of serum soluble corin was significantly higher in ischemic stroke patients (1905.03 pg/mL) than hemorrhagic stroke patients (1739.19 pg/mL) (P <0.05). No significant difference in corin was found between ischemic and hemorrhagic stroke patients in women (1352.41 pg/mL vs. 1290.24 pg/mL). The mean level of log-transformed serum soluble corin was significantly higher in ischemic stroke patients than hemorrhagic stroke patients in both men and women (all P <0.05).

Serum soluble corin in stroke subtypes

As shown in Table 3, log-transformed serum soluble corin was compared among different stroke subtypes. Among men, log-transformed serum soluble corin was significantly higher in thrombotic and embolic stroke patients than hemorrhagic stroke patients (P < 0.05), even after adjustment for age, body mass index, systolic blood pressure, triglyceride, and fasting plasma glucose. No significant difference in log-transformed serum soluble corin was observed between lacunar stroke patients and hemorrhagic stroke patients. Among women, we did not find a significantly increased level of log-transformed serum soluble corin in thrombotic or embolic stroke patients compared with hemorrhagic stroke patients. Instead, we found a significantly increased level of log-transformed serum soluble corin in lacunar stroke patients compared with hemorrhagic stroke patients (P < 0.05).

In addition, we further compared the levels of serum soluble corin among subtypes of ischemic stroke. As shown in Table 4, there were 320 (68.1%) thrombotic stroke patients, 48 (10.2%) embolic stroke patients and 102 (21.7%) lacunar stroke patients. Unadjusted analysis failed to show a significant difference in log-transformed serum soluble corin among different ischemic stroke subtypes in both men and women. However, after adjustment for the covariables, the mean level of log-transformed serum soluble corin was significantly increased in embolic stroke patients compared with other subtypes in men (P <0.05). In women, embolic stroke patients also had the highest mean level of log-transformed serum soluble corin but with no significant difference.

Results of sensitivity analysis

After excluding the stroke patients with a history of coronary heart disease, the mean level of log-transformed serum soluble corin in embolic stroke patients was the highest and significantly higher than hemorrhagic stroke patients (Table 5). Nevertheless, we failed to observe a significant difference in log-transformed serum soluble corin among subtypes of ischemic stroke (Table 6).

Discussion

This study evaluated serum soluble corin levels in patients with different types of stroke. We found that serum soluble corin levels were significantly higher in ischemic stroke patients than that in hemorrhagic stroke patients in both men and women. And it was also significantly higher in embolic stroke among ischemic stroke subtypes after adjusting for some confounding factors. The difference of ischemic stroke and hemorrhagic stroke is whether there is one or more infarctions [16]. So our results indicated that corin may play an important role in the formation of infarctions in ischemic stroke, especially in embolic stroke.

In our previous studies, serum soluble corin levels were higher in men than women regardless of ischemic or hemorrhagic stroke [10]. So we compared serum soluble corin levels among patients with different stroke types in men and women respectively. The results showed that corin levels were just higher in embolic stroke patients than that in other ischemic stroke subtypes in men but not in women. This may result from the few sample size of different stroke types in women. This is the first study to research the difference of serum soluble corin levels among stroke subtypes. It increased the possibility that corin may take part in the pathogenesis of stroke and provided a new idea for the research of stroke etiology. Furthermore, it provided a population-based evidence for clinical application of corin.

There have been reports that high B-type natriuretic peptide and midregional pro-atrial natriuretic peptide were significantly associated with a substantially increased risk of cadioembolic stroke, but the exact mechanisms were unknown [17,18]. And B-type natriuretic peptide had diagnostic value to identify cadioembolic subtype from other ischemic stroke subtypes [19,20]. As we know, corin can convert natriuretic peptides from inactive precursors to mature active forms [6,7]. In this study, we did not measure the serum level of B-type natriuretic peptide or A-type natriuretic peptide, so we did not know serum level of natriuretic peptides. In view of the relationship of corin and natriuretic peptides, it was unknown that corin directly affected stroke or indirectly affected stroke by the conversion of natriuretic peptide. Corin is a newly found protease and its biological functions are limitedly known. So the relationship between corin and cadioembolic stroke deserves further study.

We took a sensitivity analysis after excluding the patients with a previous coronary heart disease to examine the influence of coronary heart disease on corin level. Because some previous studies have proved that serum corin levels reduced in acute coronary syndrome [9]. We found that the differences of corin levels between patients of ischemic stroke subtypes and hemorrhagic stroke patients still existed while the differences among thrombolic, lacunar and embolic stroke patients did not exist. It suggested that the main source of infarction was cardiogenic because the phenomenon of elevated corin levels was disappeared after excluding the stroke patients with coronary heart disease. Soluble corin levels of embolic stroke patients were higher than patients with other ischemic stroke subtypes and were close to that in patients with coronary heart disease in previous study. We did not find a modifiable effect of coronary heart disease on the association of corin with stroke subtypes in our previous study. It indicated that corin was more closely associated with cardiovascular abnormalities than cerebral vascular abnormalities. And it can be used in the differential diagnosis of embolism from the heart.

To our knowledge, no previous study has examined the relationship between soluble corin and stroke subtypes. And our blood samples were drawn within 48 hours after the stroke onset, so we can exclude the influence of injuries resulting from cerebral ischemia or other outcomes associated with cerebral infarction, although the change of corin in different stages of ischemic stroke was unknown. Our study also has its limitations because of the small size of different types of stroke patients from a single ethnic. So the differences of corin levels among stroke subtypes need to be confirmed in other populations besides the Chinese nationality. The study was a case only study, which cannot infer a causal relationship between corin and embolic stroke. In addition, we only measured the serum corin levels, but not natriuretic peptides levels. So we did not infer that corin was directly or indirectly associated with the embolism from heart.

In conclusion, our study indicated that corin may be used in the differential diagnosis of embolism from the heart. Further study regarding to predictive and diagnostic value of corin is warranted.

Acknowledgements

We are deeply appreciative of the participants in our study and grateful to all staffs for their assistance.

Tuberculosis affects about a third of mondial population mostly resident in developing countries with inadequate health services [1]. In 1990 this disease was considered a global emergency because of the increased number of immigrants from developing countries to the industrialized ones, and the great number of HIV-infected patients [2].

When tuberculosis affects the reproductive system, it causes devastating effects, i.e. infertility due to an irreversible damage to the fallopian tubes [3,4]. Female urogenital tuberculosis often remains silent or may not present a specific symptomatology, so that its prevalence is largerly underestimated. In developing countries (such as Africa) the incidence of this disease is about 15-19 % [5], in USA, Australia and Western European countries the incidence is less than 1% [6-7].

Postmenopausal tuberculosis of the endometrium is not a common condition. It usually presents with abnormal vaginal bleeding [1,4,8]. Diagnosis is not simple [9] and it often needs laparoscopy, laparotomy or dilatation and courettage of uterine cavity (D&C). Imaging investigations are not specific. A standard anti-tubercular therapy of 6 months [10,11], is considered sufficient to obtain a complete therapeutic response, but in some cases patients need a surgical treatment [12,13].

The authors report a case of asymptomatic tubercular endometritis in a Romanian woman, in post-menopausal age, who lived in Italy for five years, from 2004 to 2009.

Case Report

A 59 years-old Romanian women with a history of HCV-related chronic liver disease, underwent abdominal ultrasound scan. She was totally asymptomatic. The patient was obese (weight: 82 kg, height:165 cm, body mass index (BMI):30,1 kg/m2 ). She referred no pregnancy in her life, menopause at the age of 50 years and no smoke or alcohol assumption. She was affected by arterial hypertension, diabetes and chronic gastritis. Her home therapy was: Felodipine 10 mg cp (1cp/die), Lansoprazole 15 mg cp (1cp/die) and Metformin 500 mg cp (1 cp twice).

The ultrasound of the upper abdomen just documented a mild liver steatosis. No abnormal sonographic appearance was identified in gallbladder, pancreas, kidneys and spleen. The trans-abdominal ultrasound of the lower abdomen showed a physiologically relaxed bladder without parietal or endoluminal alterations. Collaterally a huge hypoechoic area centrally located in the uterine shape was detected (Figure 1A).

1. 

   Figure Options

   • View in workspace
   • PowerPoint Slide

   Figure 1:

   ULTRASOUND IMAGES: A: Transabdominal ultrasound: huge hypoechoic area centrally located in the uterine shape: B-C: Transvaginal ultrasound: huge disomogeneous endometrial mass (AP diameter > 2 cm) that seems to be characterized by necrotic-colliquative material, disappearance of the line of the cavity and little demarcation with myometrium. D: Transvaginal ultrasound: Colour Doppler shows no vascularization of the ipoechoic area of the uterus.
   
   
   

Due to the patient’s globular abdomen, it was difficult to obtain further informations about the uterus by using the trans-abdominal probe. So the echographist, in agreement with the patient, decided to perform trans-vaginal (TV) ultrasound. The uterus appeared of normal dimensions, and it was median, anteversed and anteflected. There was an abnormal thickening of endometrium (AP diameter > 2 cm) that presented lack of homogeneity, disappearance of the cavity line and a little demarcation with myometrium (Figure 1B,C). With Color-Doppler, no vascularization was appreciated into the hypoechoic area of the uterus (Figure 1D). No liquid was detected in Douglas, nor adnexal diseases. The patient never had abnormal vaginal bleeding.

Because of this abnormal post-menopausal endometrial thickening, a MR scan with contrast was performed to better analyze the endometrium and the perirectal and paravaginal tissues, and to have informations about aortic, iliac and obturatory lymphonodal stations. The MR scan confirmed the relevant endometrial thickness (AP diameter > 2 cm) needing a histological characterization (Figure 2A,B).

1. 

   Figure Options

   • View in workspace
   • PowerPoint Slide

   Figure 2:

   A: T2-weighted MR scans: the uterine mass appears hyperintense and there’s no evidence of pathology involving ovaries or limphnodes. The bladder appears full and bright. B: T1-weighted MR scans: the uterine mass appears hypo-intense.
   
   
   

So the patient underwent hysteroscopy with endometrial biopsy. We don’t have hysteroscopic images because the patient performed hysteroscopy elsewhere without giving us iconographic report. The histological diagnosis was that of chronic granulomatous not necrotizing tubercular endometritis. The microbiologic cultures for alcohol-acid resistant bacilli and mycete were negative.

The patient started a standard oral treatment for tuberculosis with combination of two drugs: Rifampicin 600 mg once a day and Isoniazid 200 mg three times a day. After 40 days of therapy, the patient underwent a control TV: the huge disomogeneous and hypoechoic area, was replaced by an anechoic area because of the fluidization of the caseous material induced by the therapy (Figure 3). After this ultrasound control the patient came back to Romania. Then she referred she continued the anti-tubercular therapy for 6 months with a complete remission of the endometrial tuberculosis. No ultrasound images are available of the complete remission of the pathology.

1. 

   Figure Options

   • View in workspace
   • PowerPoint Slide

   Figure 3:

   Transvaginal ultrasound 40th days after the start of the therapy: fluidization of the caseous material.
   
   
   

Discussion

In 1990, a total of 3,8 million cases of tuberculosis were notified to WHO, 49% of them were from sud-est of Asia. An association with HIV infection was revealed [14]. The main pathological localization of Mycobacterium tuberculosis is the respiratory tract; however, all the organs of the body can be reached by this infective agent: bones, arteries, kidneys, liver, spleen, prostate gland, epididymides, uterus, gastro-enteric tract, adrenogenital glands, subarachnoid space, peritoneum, pericardium and lymph nodes [9,15].

Primary genital tuberculosis is extremely rare, whereas it is almost always secondary to tuberculosis infection elsewhere in the body. Genital tuberculosis is the most common manifestation of extra-pulmonary tuberculosis [8,9,16], representing the 15-19% of extra pulmonary tubercular cases [5]. The global prevalence of genital tuberculosis has increased from 22 million cases in 1995 to 1.86 billion cases in 2005, with a 5-10% rate of infertility [17]. It is usually misunderstood by health-care providers, but is an important cause of significant morbidity, with short- and long-term sequelae for the affected women [18].

The extra-pulmonary foci are probably colonized by hematogenous spread of the bacilli. The main genital localizations of Mycobacterium tuberculosis are the fallopian tubes (90-100 % of cases) [19], then it spreads towards uterus (50-60%), ovaries (20-30%), cervix (5-15%), and vagina (1 %) [10,20,21].

Patients with genital tuberculosis are usually young women with a story of infertility [8], chronic pelvic disease [22], or ectopic pregnancy [23]. Genital tuberculosis is usually asymptomatic. The main causes of a gynecological consultations are infertility (44%), pelvic pain (25%), abnormal vaginal bleeding (18%), amenorrhea (5%) and leukorrhea (4%). Patients can rarely present with abdominal mass, ascites or a tubo-ovarian abscess [24].

Postmenopausal tuberculosis of the endometrium is not a common condition. It usually presents with abnormal vaginal bleeding [1,4,8,24], but it may also present without specific symptoms. This is why there’s usually a delay in diagnosis and in therapeutic interventions. Many authors explain the low incidence of this disease in post-menopausal age, assessing that atrophic endometrium represents a poor ground for mycobacterial growth [1,25]. Abnormal vaginal bleeding represents the 5% of causes of a gynecological consultation in post-menopausal age [26]. Differential diagnosis to consider may be endometrial atrophy (59%), endometrial polyps (12%), endometrial cancer (10%), pyometra (13,6%), endometrial hyperplasia, hormonal response, cervical cancer, other infections [8,27,28]. In the differential diagnosis of an ovarian tumor, an endometrial cancer and ascites, genital tuberculosis should always be considered [13,29,30].

Diagnosis of genitourinary tuberculosis is not simple [9] and often may need laparoscopy, laparotomy or dilatation and curettage of uterine cavity (D&C). Urine culture is not useful for diagnosis. Imaging investigations are not specific. Hysterosalpingography may show calcified pelvic lymphadenopathy and adhesions in the uterine cavity with some typical aspect, but it is not very specific nor sensible [20]. Dadhwal et al. [18], reported that the clinician should be aware that isolation of Mycobacterium Tuberculosis requires special methods and this diagnosis should be considered while dealing with patients born in countries with high prevalence of Tuberculosis.

Definitive diagnosis requires the identification of Mycobacterium tuberculosis either by direct microscopic examination or after culturing pathological specimens. Tangappah [3], reported that in the presence of positive molecular investigation with PCR an infertile woman should be considered as having genital tuberculosis and should be treated. Concerns remain about the high false-negative results with PCR techniques. Besides the long period required for their culture, in paucibacillary endometrial samples, acid-fast bacilli smears are almost always negative [31].

For genital tuberculosis, treatment guidelines recommend a standard anti-tubercular therapy of 6 months, providing that pyrazinamide is included for the first two months of treatment and that the organism is susceptible [11]. The main anti-tubercular drugs are: Streptomycin, Isoniazid, Ethambutol, Rifampicin, and Pyrazinamide [10].

Surgical treatment may be considered in case of persistence of adnexal mass (especially the cold abscess) after medical approach, recurrence of endometrial tuberculosis after a year of treatment, persistence of pelvic pain after 3 months of medical therapy, persistence of vaginal bleeding, fistulas that do not resolve [12,13]. The surgical treatment should be carried out at least 6 weeks after initiation of the pharmacological treatment, because this reduces the risk of intraoperative complications and facilitates the surgical procedure itself [1].

Conclusions

Tubercular endometritis is a severe condition that may affect both pre- and post-menopausal woman and may present both with or without symptoms. Diagnosis may be difficult and requires laparoscopy, laparotomy or D&C. Ultrasound examination may detect a mass of the lower abdomen but it does not provide information about the nature of the mass itself.

A standard anti-tubercular therapy of 6 months is considered adequate to obtain a complete therapeutical response; however, in some cases a surgical approach may be necessary. Progress should be done in diagnostic strategies of this disease, to avoid the progression of the pathology in postmenopausal age, and to correct the damages caused at the reproductive system in fertile age.

This paper shows that postmenopausal tubercular endometritis may be asymptomatic and even if it is not a common condition, it has always to be considered in differential diagnosis of an ovarian tumor and an endometrial cancer.