Patients with a sharp and unbearable recurrent/constant chest and/or back pain with a duration of at least 5 min each time were recruited. A total of 70 patients with AAD and 12 patients with aortic intramural hematoma (IMH) were selected in a consecutive manner at the Zhongshan Hospital Fudan University (Shanghai, China), and were matched with 30 healthy volunteers. Patients with trauma, syphilis, a creatinine level of >355 µmol/l, hemoglobin level of <60 g/l or systolic blood pressure of <90 mmHg prior to treatment were excluded. The diagnosis of AAD or IMH was confirmed using MSCTA. In cases with AAD, an intimal flap showing a tear of the intimal layer was observed. Unlike the false lumen in AAD cases, the crescent-shaped area observed in IMH patients was not enhanced subsequent to contrast agent administration, and no intimal tear was observed on contrast-enhanced CT scans (16). All study subjects provided their informed consent for participation in the study. The study was approved by the Ethics Committee of Zhongshan Hospital Fudan University. For each study subject, whole blood samples were immediately collected in BD Vacutainer SST tubes (BD Diagnostics, Plymouth, UK) following admission and were centrifuged at 2,770 × g for 10 min at room temperature. The serum was frozen and stored in aliquots at −80°C until required for analysis.

The Stanford system was used to describe the aortic dissection in the patients (17). If the dissection involved the ascending aorta and/or aortic arch, then patients were categorized as Stanford Type A. The type A patients generally require surgical treatment, such as the Bentall procedure which is performed to prevent the aorta from rupturing and causing a fatal hemorrhage. This surgical procedure involves replacing certain defective parts of the aorta, such as the valve or the upper part (also known as the ascending aorta), with a graft.

Serum levels of lumican were measured using ELISA kits from Cusabio Biotech Co., Ltd. (Wuhan, China; cat. no. CSB-E09797h). Briefly, each sample (100 µl) was added to a 96-well microplate and incubated for 2 h at 37°C. The liquid was then removed, and the biotin-antibody working solution (100 µl) was added to each well and incubated for 1 h at 37°C, after which the liquid was aspirated, and the wells were washed with 200 µl wash buffer three times. Horseradish peroxidase-avidin working solution (100 µl) was added to each well and incubated for 1 h at 37°C. Subsequent to five washes, 90 µl 3,3′,5,5′-tetramethylbenzidine was added to each well and incubated for 30 min at 37°C. The enzymatic reaction was terminated by adding the stop solution. Thereafter, the absorbance was measured at 450 nm using a Bio-Rad model 680 microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

To confirm the localization of the lumican protein in the human aortic sections, immunohistochemical staining and western blot analysis were performed. The aortic sections were obtained from 7 patients that had previously undergone the Bentall procedure and from 2 heart transplant donors, after obtaining patient consent.

Immunohistochemical assay was performed according to the protocol of the Vectastain ABC kit (Vector Laboratories, Burlingame, CA, USA). Briefly, aortic tissue samples were fixed with freshly-prepared 4% paraformaldehyde and then dehydrated. The tissues were then embedded in molten paraffin and stored at room temperature until ready for sectioning. A microtome (RM2235; Leica Biosystems, Wetzlar, Germany) was used to cut the embedded tissue into 5–20-µm sections, which were then floated in a 50°C water bath. Next, the sections were mounted onto gelatin coated slides, allowed dry overnight and stored at room temperature until ready for staining. The sections were then deparaffinized, rehydrated and washed twice for 10 min with 1% animal serum in phosphate-buffered saline (PBS) with 0.4% Triton X-100 (PBS-T). Any non-specific binding was blocked by incubating the tissue sections with 5% serum in PBS-T for 30 min at room temperature, followed by incubation with polyclonal rabbit anti-human lumican antibody (Novus Biologicals, Littleton, CO, USA; cat. no. NBP1-87726; dilution, 1:200). Subsequently, the sections were incubated with ABC-HRP reagent (part of the Vectastain ABC kit; formed by mixing avidin with biotinylated horseradish peroxidase) for 1 h at room temperature and then washed twice in PBS for 10 min each time. The tissue sections were then counterstained with hematoxylin for 2 min and dehydrated, and mounting media (neutral balsam; Sangon Biotech Co. Ltd., Shanghai, China) was added.

For western blot analysis, the acute aortic dissection samples were lysed with lysis buffer (Sigma-Aldrich, St. Louis, MO, USA), total protein was extracted from the tissue samples, and the protein concentrations were determined using a protein assay kit (Bio-Rad Laboratories, Inc.). Next, 30 µg protein was subjected to 10% SDS-PAGE and transferred to nitrocellulose membranes (Bio-Rad Laboratories, Inc.). The membranes were blocked with 5% non-fat milk for 1.5 h. Subsequently, the samples were incubated with polyclonal rabbit anti-human lumican antibody (Abcam, Cambridge, UK; 1 µg/ml; cat. no. ab98067) for the detection of lumican, followed by horseradish peroxidase-conjugated secondary antibodies (Thermo Fisher Scientific, Inc., Waltham, MA, USA; cat. no. A16104SAMPLE; dilution, 1:1,000). After incubation with secondary antibody, the proteins were detected with enhanced chemiluminescence reagents (Pierce Biotechnology, Rockford, IL, USA).

All patients and healthy volunteers underwent MSCTA imaging. Whole blood samples were collected immediately after admission; however, due to practical difficulties, MSCTA imaging was performed at a later time (<3 h after admission). All CT examinations were performed using a Somatom Definition AS MSCT system (Siemens Medical Solutions, Forchheim, Germany) using a standard clinical protocol. Scanning parameters were as follows: Voltage, 140 kV; effective mAs value per rotation, 400 mAs; gantry rotation time, 0.33-msec; collimation with z-flying focal spot for each detector, 128×0.6 mm. The pitch varied between 0.2 and 0.43. For dose reduction, an automatic tube current modulation algorithm was used. A standard volume of 85 ml contrast material (Isovue 370; Bracco Diagnostics, Inc., Seattle, WA, USA) was administered at 3 ml/sec followed by a 50-ml saline flush. The automatic detection of peak enhancement in the descending aorta with a detection threshold of 120 Hounsfield units was used, while the helical scan was initiated automatically after the bolus detection. CT data sets were reconstructed at a slice thickness of 1 mm.

Reconstructed images were transferred to a remote workstation and a cardiac function analysis software package (GE/Siemens Workstation: ADW 4.2, General Electric, Schenectady, NY, USA; Syngo MMWP VE36A, Siemens Healthcare, Erlangen, Germany) was used. The CT data sets were analyzed in a consensus reading by two experienced cardiovascular radiologists.

Due to the complexity and inaccuracy of direct measurements, the vertical range of the false lumen was measured and we determined whether vital vessels were affected to estimate its surface area and the severity of AAD. Thus, a ‘SCORE X, RANGE Y’ system was designed to measure the number of groups of affected vital arteries and the vertical range of the false lumen. This system was designed by our group, based on the fact that the extent of the dissection and branch-vessel involvement may increase the morbidity and mortality (18).

SCORE grading depended on the number of affected vital vessels. The different groups of affected vital vessels were as follows: Group 1, including the brachiocephalic trunk, left common carotid artery and left subclavian artery; group 2, including the celiac trunk and superior mesenteric artery; and group 3, including the bilateral renal artery and bilateral iliac artery. If the affected vessel fell within one of the groups, the SCORE was marked as ‘1’. If the affected vessel fell within two of the groups, the SCORE was marked as ‘2’. If no vital vessels were affected, the SCORE was marked as ‘0’. The maximal and minimal SCORE were 3 and 0, respectively.

RANGE grading depended on the number of aortic segments that were affected by the false lumen. The aorta was divided into the following segments: i) Ascending to arch; ii) arch to celiac trunk level; iii) celiac trunk level to the level of the renal arteries; and iv) level of the renal arteries to the iliac arteries. If one segment was affected, RANGE was marked as ‘1’. If two segments were affected, RANGE was marked as ‘2’, If three segments were affected, RANGE was marked as ‘3’. If all segments were affected, RANGE was marked as ‘4’.

Data were collected and inputted in Excel (Microsoft Corporation, Redmond, WA, USA). All statistical analyses were performed using SPSS software, version 17.0 (SPSS, Inc., Chicago, IL, USA) and GraphPad Prism 5 software (GraphPad Software, Inc., San Diego, CA, USA), and the results were presented as mean ± standard deviation. A comparative analysis of multiple groups was performed with a one-way analysis of variance or Mann-Whitney/Kruskal-Wallis test. Spearman's correlation coefficients were used to quantify the correlations. Bivariate correlation analyses were used to present the results, and P<0.05 was considered to indicate a statistically significant difference.

In total, 70 patients with AAD, 12 patients with IMH and 30 healthy volunteers were enrolled in the present study. The clinical features of all subjects are summarized in Table I. No statistically significant differences in age distribution, gender composition and past medical history (hypertension and Marfan syndrome, which is a genetic disorder of connective tissue resulting in defects of the heart valves and aorta, and occasionally aortic dissection) were observed among the three groups (P=0.148, 0.275, 0.620 and 0.252, respectively). Furthermore, there was no differences in the time from onset to admission between the AAD and IMH groups (P=0.331). However, systolic blood pressure was increased in the AAD and IMH patients compared with the healthy volunteers (P=0.002).

As shown in Fig. 1A, a statistically significant difference in serum lumican levels was detected among AAD patients (2.32±4.29 ng/ml), IMH patients (0.72±0.32 ng/ml) and healthy volunteers (0.85±0.53 ng/ml; P=0.003). Furthermore, the association between the level of serum lumican and the time from symptom onset (Fig. 1B) in AAD patients was assessed, and a significant difference was observed between patients presenting at the hospital within 12–72 h (3.21±5.57 ng/ml) and those presenting within <12 h (1.13±1.10 ng/ml) or >72 h (1.65±1.99 ng/ml; P=0.005) from the symptom onset.

In total, 8 patients presented at >72 h, including 4 patients with Stanford Type A disease, of which 1 patient underwent the Bentall procedure, and 4 patients with Stanford Type B disease, of which 1 patient received endovascular graft exclusion treatment. Among the 24 patients that presented at <12 h, 7/14 Type A and 7/10 Type B patients were treated surgically. As the serum lumican levels are significantly associated with the time from symptom onset, further analysis was only performed in 38 patients that presented within 12–72 h after an episode of chest/back pain (of these, 9/25 Type A and 8/13 Type B patients were treated surgically). In total, 9 subjects were excluded as their diagnosis was not confirmed by MSCTA or MSCTA data acquisition was obtained at >3 h following admission. Among the remaining 29 AAD patients, 7/16 Stanford Type A patients had undergone the Bentall procedure, and 8/13 Stanford Type B patients had received endovascular graft exclusion treatment.

As shown in Fig. 2A, the lumican levels in patients with SCORE values of 0, 1, 2 and 3 were 1.15±1.61, 2.31±2.61, 1.64±1.57 and 6.63±8.43 ng/ml, respectively. The Spearman's rho correlation coefficient between lumican levels and SCORE was 0.373 (P=0.046), which indicated that the level of lumican was correlated with the number of affected vital vessels. In patients with RANGE values of 1, 2, 3 and 4 (Fig. 2B), the serum lumican levels were 0.43±0.14, 4.46±4.41, 1.45±1.40 and 5.57±7.58 ng/ml, respectively. In addition, the correlation coefficient between lumican levels and RANGE was 0.468 (P=0.010), which indicated that lumican levels were correlated with the range of the false lumen.

Among the 29 eligible AAD patients presenting within 12–72 h after onset, 7 Standford Type A patients had received the Bentall procedure. Immunohistochemical staining and western blot analysis of lumican were performed in the aortic sections from the 7 patients that had previously undergone the Bentall procedure and from 2 heart transplant donors. Western blot analysis (Fig. 3A) and immunohistochemical staining (Fig. 3B and C) were performed to detect the expression of lumican in the aortic sections of AAD patients and donors.

Images demonstrating the expression of lumican from the aortic sections of a 37-year-old male patient are shown in Fig. 3B. This patient had no history of hypertension or family history of aortic dissection, and presented with sharp chest pain for 72 h, following which he underwent a Bentall procedure and survived. The evaluation for this patient was SCORE 3 and RANGE 4, and the serum lumican level was 12.54 ng/ml. Fig. 3B shows the expression of lumican (brown) in the aortic medial layer (magnification, ×200). The expression of lumican in the aortic medial layer of a 24-year old male heart transplant donor is shown in Fig. 3C (magnification, ×200). Although lumican was expressed in the aortic medial layer of AAD patients and healthy donors, the serum levels of lumican were elevated in AAD patients, but not in the healthy donors, which may reflect our hypothesis that the medial layer released more lumican into the circulation via the blood flow from the intimal tear.