If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
In Japan, ventricular assist devices (VADs) have been used for patients with severe heart failure as a bridge to transplantation (BTT) since 1992. However, it was not until 1997, when the Organ Transplant Law was enacted, that medical devices received approval by the national health insurance system for that use. To encourage research and development of innovative medical devices, the Pharmaceuticals and Medical Devices Agency has established a public–private partnership in collaboration with academic societies, hospitals and manufacturers.
Methods
The Japanese registry for Mechanically Assisted Circulatory Support (J-MACS) is a prospective registry designed to be harmonized with the Interagency Registry of Mechanically Assisted Circulatory Support (INTERMACS). Participation in J-MACS is mandatory for device manufacturers to meet the conditions of approval as well as for hospitals to obtain authorization for reimbursement from the national health insurance system.
Results
From June 2010 to April 2015, 476 patients were registered at 31 hospitals. Of these, analysis of primary VAD patients (n = 332) revealed that their overall 360-day survival was 91% (implantable 93%, extracorporeal 84%).
Conclusions
This initial report from J-MACS focuses on patients’ demographics, device types, survival, competing outcomes, adverse events and successful examples of system failure detection.
In Japan, the Organ Transplant Law, enacted in 1997, paved the way for patients with severe heart failure to be officially listed for heart transplantation.
Although ventricular assist devices (VADs) continue to undergo evolutionary refinements, their use and development remain challenging for both health-care providers and device manufacturers, due to the typically long waiting period for qualified patients to undergo heart transplantation.
Domestic and foreign trends in the prevalence of heart failure and the necessity of next-generation artificial hearts: a survey by the Working Group on Establishment of Assessment Guidelines for Next-generation Artificial Heart Systems.
was established in 2009 and is the first national registry designed to be harmonized with the Interagency Registry of Mechanically Assisted Circulatory Support (INTERMACS).
The primary aims of J-MACS are to utilize VAD data collected by the J-MACS registry to improve clinical assessments and management, as well as treatments and related technologies for patients with severe heart failure. In addition, by clarifying associated risks and benefits, the data obtained are used to establish appropriate safety measures and promote development of next generation technology.
J-MACS design, structure and organization
J-MACS is a public–private partnership that includes 7 different academic societies, as well as participating hospitals and relevant VAD manufacturers, and is funded by the Pharmaceuticals and Medical Devices Agency (PMDA) of Japan (Figure 1). A Steering Committee (SC) and an Operating Committee (OC) have been established for management of J-MACS, with the SC consisting of representatives from the academic societies, along with experts from the participating hospitals and manufacturer associations, whereas the OC consists of a principal investigator (PI) and co-PIs, as well as experts from the participating hospitals and VAD manufacturers. In addition, the Observational Study Monitoring Board (OSMB) and Adverse Event and Adjudication Committee (AEAC) were established as independent and impartial organizations. The OSMB oversees the registry, whereas the AEAC investigates major adverse events (device malfunction, bleeding, neurologic dysfunction, infection) and death in registered cases.
Figure 1Organization of Japanese registry for Mechanically Assisted Circulatory Support (J-MACS). PAL, Pharmaceutical Affairs Law, AE, adverse events; HBD, United States–Japan Medical Device Harmonization by Doing; IMACS, International registry for Mechanically Assisted Circulatory Support. Academic Societies: Japanese Society for Artificial Organs; Japanese Association for Clinical Ventricular Assist Systems; Japanese Association for Thoracic Surgery; Japanese Society for Cardiovascular Surgery; Japanese Circulation Society; Japanese College of Cardiology; and Japanese Heart Failure Society.
J-MACS routinely and/or immediately provides prospectively obtained clinical data, including patient demographics, operative surgery information, post-operative follow-up findings and adverse events, via the J-MACS web-based data entry system. It should be noted that VAD manufacturers are allowed to utilize J-MACS data for reporting medical device malfunctions and conducting post-marketing surveillance, as required under the Pharmaceutical Affairs Law (PAL) and Act on Securing Quality, Efficacy and Safety of Pharmaceuticals, Medical Devices, Regenerative and Cellular Therapy Products, Gene Therapy Products, and Cosmetics of Japan.
The present study design, including data elements, follow-up schedule and adverse event definitions, was implemented so as to harmonize with INTERMACS
U.S.-Japan Harmonization by Data (HBD)—initial experience with a post-marketing follow-up study of mechanical circulatory assist devices (J-MACS), and ongoing collaboration with academic research consortiums.
Pharmaceut Med Dev Regul Sci [in Japanese].2012; 43: 339-346
Requirement for site participation and auditing system
Membership in J-MACS is one of the essential conditions for an authorized hospital to use an implantable LVAD and receive national medical insurance coverage. As of January 2015, 40 hospitals in Japan were authorized by relevant academic societies to utilize an implantable VAD for bridge to transplantation (BTT) in qualified patients. A PI has been established at each participating site and is responsible for data entry via the J-MACS web-based data entry system.
With an aim to enhance the quality of the registry, J-MACS conducts site visits for auditing. The audit committee consists of supervisory physicians experienced in VAD therapy and works with the data-coordinating center (DCC). Members of the committee periodically visit selected hospitals to examine data integrity, including confirmation of entered data and documentary records. The DCC also urges participating hospitals to input all eligible patient data based on reports (e.g., number of implanted devices) provided by VAD manufacturers. The committee makes a report of the auditing results and submits that to the OC. After confirming the report, the OC informs the SC regarding its contents.
Device types
Table 1 lists the VADs included in the J-MACS registry. When the J-MACS system was launched in June 2010, an extracorporeal pulsatile-flow device was the only type available for BTT. Thereafter, 2 implantable continuous-flow devices were approved for use in Japan in December 2010.
Patient inclusion criteria for J-MACS are as follows: (1) a durable VAD approved for use in Japan, as noted in Table 1, has been implanted; (2) the VAD implantation was performed at a hospital authorized by the cooperating academic societies; and (3) an informed consent form for inclusion in the registry has been signed by the patient and/or a family member. In addition, patients who have applied to receive an extracorporeal type of LVAD at an authorized hospital as BTT and provided informed consent for inclusion are also included in the J-MACS registry.
Between June 2010 and April 2015, 476 patients from 31 participating hospitals were enrolled. In the early stage, only patients with an extracorporeal type of LVAD were included. However, since the introduction of implantable continuous-flow devices in December 2010, the number of enrolled patients has dramatically increased (Figure 2). For this first report, we analyzed 332 patients who met all of the selection criteria (primary LVAD population), which are as follows: (1) assisted only by an LVAD; (2) never used any VAD before time of enrollment; and (3) >18 years of age. The cut-off date for data collection was July 8, 2015. The rates for compliance of data entry regarding follow-up examinations were 92% at 1 month, 92% at 12 months and 91% at 24 months.
Male gender accounted for 79% of the patients (implantable 81%, extracorporeal 73%) (Table 2). The majority of patients with an implantable type device ranged from 30 to 59 years old, whereas younger patients more often received an extracorporeal type. The primary disease in the majority of patients was dilated cardiomyopathy.
Table 2Patients’ Demographics (J-MACS: June 2010 to April 2015)
Implantable (%) (n = 259)
Extracorporeal (%) (n = 73)
Total (%) (n = 332)
Gender
Female
49 (19)
20 (27)
69 (21)
Male
210 (81)
53 (73)
263 (79)
Age range at implant (years)
10 to 19
6 (2)
1 (1)
7 (2)
20 to 29
32 (12)
18 (25)
50 (15)
30 to 39
59 (23)
20 (27)
79 (24)
40 to 49
70 (27)
17 (23)
87 (26)
50 to 59
72 (28)
14 (19)
86 (26)
60 to 69
19 (7)
3 (4)
22 (7)
70 to 79
1 (0)
0 (0)
1 (0)
Mean ± SD
43.4 ± 12.2
38.9 ± 12.1
42.4 ± 12.3
Height (cm)
167.7 ± 8.0
168.0 ± 7.9
166.5 ± 8.5
Body mass index (kg/m2)
20.7 ± 3.6
20.7 ± 3.3
20.8 ± 4.7
Body surface area (m2)
1.65 ± 0.18
1.66 ± 0.16
1.63 ± 0.22
Pre-implant cardiac diagnosis (primary)
Congenital heart disease
4 (2)
1 (1)
5 (2)
Coronary artery disease
26 (10)
10 (14)
36 (11)
Hypertrophic cardiomyopathy
32 (12)
7 (10)
39 (12)
Valvular heart disease
0 (0)
1 (0)
1 (0)
Dilated cardiomyopathy
184 (71)
52 (71)
236 (71)
Others
13 (5)
2 (3)
15 (5)
J-MACS, Japanese registry for Mechanically Assisted Circulatory Support.
The main strategy for receiving an implantable VAD in the majority of patients was for BTT, with 67% with an implantable type listed with the Japan Organ Transplant Network as a candidate for heart transplantation (Table 3). In contrast, 39% of patients with an extracorporeal type device received it as a BTT, with only 8% listed in the registry as heart transplant candidates. Nearly half (49%) of patients with an extracorporeal device were in critical cardiogenic shock (INTERMACS patient profile Level 1), whereas the majority of those with an implantable device were Level 2 (53%) or 3 (41%).
Table 3Pre-implant Patient Profiles and Device Atrategies (J-MACS: June 2010 to April 2015)
Actuarial survival rate, determined using the Kaplan–Meier method, for the entire cohort was 91.2% at 360 days and 88.7% at 720 days after implantation (Figure 3). Patients are censored at the time of transplantation, device explantation (because of recovery or switch to an implantable type in those with the extracorporeal type) or at time of the last observation. Actuarial survival rates for patients with the implantable and extracorporeal types were 92.6% and 84.0%, respectively, at 360 days, and 89.8% and 84.0%, respectively, at 720 days (Figure 4). When stratified by INTERMACS patient profile, 360-day survival was 87.4% for Level 1, 92.4% for Level 2 and 90.8% for Level 3. Patients at Level 1 showed significantly worse survival compared with those at Level 2 or 3 (Figure 5).
Figure 3Actuarial survival for entire primary LVAD population registered with J-MACS (J-MACS: June 2010 to April 2015).
Figure 4Actuarial survival stratified by implantable and extracorporeal type for primary LVAD population (implantable/extracorporeal) (J-MACS: June 2010 to April 2015).
The likelihood of BTT patients to undergo heart transplantation was 1.0% within 360 days, 7.8% within 720 days and 38.8% within 1,080 days in the implantable group (Figure 6), as compared with 0%, 1.6% and 13.7%, respectively, in the extracorporeal group (Figure 7). When patients with an extracorporeal type device as BTT received official approval to become a candidate for heart transplantation, the majority were switched to an implantable type. Approximately 63.8% of patients with an extracorporeal type underwent explantation within 360 days, mainly because of a switch to an implantable type due to the aforementioned reasons (Figure 7). Death within 1,080 days after receiving the implantable type occurred in 16.8% of those cases.
Figure 6Analysis of competing outcomes after implantation of an implantable type device in the primary LVAD population. Death: died with device; Explanted: device explantation performed; Alive: device in place. At all points in time, the sum of the probabilities of each outcome event totaled 100% (J-MACS: June 2010 to April 2015; implantable type: n = 259).
Figure 7Analysis of competing outcomes after implantation of an extracorporeal type device in the primary LVAD population. Death: died with device; Explanted: device explantation performed; Alive: device in place. At all points in time, the sum of the probabilities of each outcome event totaled 100% (J-MACS: June 2010 to April 2015; extracorporeal type: n = 73).
The primary causes of death are listed in Table 4. Infections, neurologic events and multi-organ failures were the major events occurring in patients with an implantable device.
Table 4Primary Causes of Death (J-MACS: April 2010 to June 2015)
Implantable (n = 23)
Extracorporeal (n = 8)
Primary cause of death
No.
%
No.
%
Infection
7
30.4
6
75.0
Neurologic event
7
30.4
0
0.0
Multi-organ failure
3
13.0
1
12.5
Right ventricular failure
2
8.7
0
0.0
Bleeding
1
4.4
0
0.0
Device malfunction
1
4.4
0
0.0
Others
2
8.7
1
12.5
Total
23
100.0
8
(100.0)
J-MACS, Japanese registry for Mechanically Assisted Circulatory Support.
The numbers of patients affected by each of the 4 types of major adverse events, infection, neurologic dysfunction, device malfunction and bleeding, before and 30 days after implantation, are shown in Table 5. Infection was the most common adverse event seen in the patients regardless of device type, with device-related infection seen in 22 cases (18%) with implantable type and 10 (32%) with extracorporeal type devices. In those with implantable types, infection of the cable exit site was found to be the primary cause of infection. Neurologic dysfunction was the second most common adverse event seen in patients with both types of devices, and its diagnosis was obtained based on brain computed tomography findings and clinical course at each hospital. Specifically, hemorrhagic stroke, ischemic stroke and transient ischemic attack (TIA) were seen in 35 (40%), 27 (31%) and 21 (24%), respectively, of the 87 patients with an implantable type, and 9 (41%), 5 (23%) and 5 (23%), respectively, of those with the extracorporeal type. As for device malfunction, the rate was higher in extracorporeal type cases when compared with implantable types. In the registry, a “pump exchange due to thrombus formation” event is reported as a device malfunction, as is malfunction of the device-line in an implantable device. Regarding bleeding, many of the cases were reported to have a surgical cause and required an operation, whereas gastrointestinal bleeding was not so frequent in patients with either type of device. Most device-related infections occurred after 30 days in both types and hemorrhagic stroke was also more frequent after 30 days with each type. In contrast, bleeding was more frequent prior to 30 days in cases with the implantable devices. Rates for major adverse events within the first 12 months after implantation are presented in Table 6. There was an approximately 2-fold greater number of adverse events in patients with an extracorporeal vs implantable type device.
Table 5Adverse Events (J-MACS: April 2010 to June 2015)
Adverse event
Implantable (n = 259)
Extracorporeal (n = 73)
≤30 days
>30 days
Total
≤30 days
>30 days
Total
Infection
32
90
122
10
21
31
Device-related
1
21
22
0
10
10
Non–device-related
31
69
100
10
11
21
Neurologic dysfunction
24
63
87
9
13
22
Hemorrhagic stroke
5
30
35
1
8
9
Ischemic stroke
9
18
27
4
1
5
TIA
8
13
21
2
3
5
Others
2
2
4
2
1
3
Device malfunction
16
64
80
19
18
37
Bleeding
34
19
53
9
9
18
Non-GI bleeding
33
16
49
8
9
17
GI bleeding
1
3
4
1
0
1
Values indicate number of patients affected. GI, gastrointestinal; TIA, transient ischemic attack.
J-MACS Research Group: Post-approval study of a highly pulsed, low-shear-rate, continuous-flow, left ventricular assist device, EVAHEART: a Japanese multicenter study using J-MACS.
J-MACS Adverse Event and Adjudication committee. The first report on adverse event for mechanically circulatory support. December 2012. http://www.pmda.go.jp/files/000147200.pdf/. Accessed June 1, 2016.
in their observations. Notably, thrombosis formation of an inflow cannula was detected by reviewing the high rate of ischemic stroke in a single device. According to the results of their reviews, the AEAC gives advice to the appropriate regulatory agency. Fortunately, in cooperation with relevant academic societies, VAD manufacturers have promptly improved the design of their devices based on that information.
This is the first report of the Japanese registry for mechanical circulatory support, J-MACS, in which we have focused on patients’ demographics, device type, survival, competing outcomes and adverse events. Thanks to the leadership of the International Society for Heart and Lung Transplantation (ISHLT) in establishing the International registry for Mechanically Assisted Circulatory Support (IMACS),
Our investigation of acquired data shows that Japanese patients are experiencing longer waiting times for heart transplantation while being supported by a primary LVAD compared with patients in other countries. According to our analysis of competing outcomes with implantable type devices, the number of transplant cases was clearly increased in cases supported from 720 to 1,080 days. In spite of such an extended time with LVAD support, the results of heart transplantation in these BTT cases are good, as approximately 90% survive for >10 years after transplant.
Thus, many patients in Japan with severe heart failure express their desire to be listed as a candidate for heart transplantation and undergo implantation of an implantable LVAD as BTT. Our detailed examination of those cases highlight the importance of exploring better ways to control infection, neurologic dysfunction and bleeding to improve patient quality of life. Furthermore, alternative options for patients who are not candidates for heart transplantation are needed due to the good quality of life offered by use of an implantable LVAD.
With regard to support duration, Japanese patients with an LVAD as BTT may be comparable to patients in other countries receiving LVAD support as destination therapy. However, a randomized trial of LVADs is an enormously complex and difficult task, especially when undertaken for an extended trial period.
Thus, prospective registries that contain high-quality data collection are becoming more important for regulatory agencies in this era of real-world evidence.
We believe that our database is a valuable source of safety information, especially for long-term-use cases. On the other hand, this initial investigation of available data has also revealed some local patterns specific to Japan. For example, some patients initially received an extracorporeal device before being switched to an implantable one. Such cases are a reflection of the reimbursement policy of the Japanese national health insurance system, which limits use of implantable devices to those who have been officially “listed” or “approved” as heart transplant candidates.
Among the cases analyzed, device malfunctions, such as thrombus formation, were reported as ischemic stroke related to neurologic dysfunction by the AEAC. Careful assessment of adverse events is crucial to clarify important issues and points that require improvement and for reducing the number of similar events in the future. Furthermore, a quick response to the cause of an adverse event is important for providing an opportunity to effectively treat affected patients who need a new device.
Presently, we are continuing analyses of our accumulated data in greater detail, including quality of care and device durability, with special focus on patients who receive support for longer periods. These findings will be reported in the near future.
Conclusions
J-MACS is the first internationally harmonized registry for mechanical circulatory assist devices in Japan. Presently, the 1-year survival rate for patients with an implantable LVAD is approximately 90%. When stratified by INTERMACS patient profile, survival in the present study after 360 days was 87.4% for Level 1, 92.4% for Level 2 and 90.8% for Level 3. For patients who have received implantable LVAD support for BTT, the likelihood of transplantation within 720 days is presently 7.8%. Detection of device failures and adverse events by the AEAC has facilitated prompt actions by regulators as well as manufacturers.
Disclosure statement
H.O., M.T., M.H. and S.Y. are employees of Terumo Heart KK, Century Medical, Inc., Nipro Corporation and Sun Medical Technology Research Corp., respectively. The remaining authors have no conflicts of interest to disclose. The J-MACS registry is a pilot project of the PMDA established for gathering and evaluating data regarding the use of medical devices. It is funded by post-marketing safety measure contributions from authorized marketing holders and also sponsored by Terumo Heart K.K, Century Medical, Inc., Nipro Corporation and Sun Medical Technology Research Corp. for use in post-marketing surveillance. These companies had no influence on the design of the present study, nor on data collection, analysis and interpretation. This study was supported by the Intramural Research Fund for Cardiovascular Diseases of the National Cerebral and Cardiovascular Center (Grant 25-4-1 to T.N.) and by JSPS KAKENHI (Grant 25350585 to K.S.). The authors thank all of the participating hospitals (Appendix 1), J-MACS project members (Appendix 2) and cooperating academic societies (Appendix 3) for their advice and help with data collection. We also thank Dr. Bram D. Zuckerman and Dr. Eric A. Chen for their support as part of the United States–Japan Harmonization by Doing program. We are also grateful for the support and encouragement of Dr. James K. Kirklin, Dr. David C. Naftel and the INTERMACS investigators.
Appendix 1. Participating hospitals
Akita University Hospital, Chiba University Hospital, Dokkyo Medical University Hospital, Ehime University Hospital, Fukui Cardiovascular Center, Gunma Cardiovascular Center, Hokkaido University Hospital, Hyogo College of Medicine Hospital, Japanese Red Cross Nagoya Daini Hospital, Jichi Medical University Hospital, Kinki University Hospital, Kitasato University Hospital, Kurume University Hospital, Kyoto University Hospital, Kyushu University Hospital, Mitsui Memorial Hospital, Nagasaki University Hospital, Nagoya Tokushukai General Hospital, Nagoya University Hospital, National Cerebral and Cardiovascular Center, Nihon University Itabashi Hospital, Oita University Hospital, Okinawa Prefectural Nanbu Medical Center & Children's Medical Center, Osaka University Hospital, Saga University Hospital, Saitama Cardiovascular and Respiratory Center, Saitama Medical Center Jichi Medical University, Saitama Medical University International Medical Center, Saku Central Hospital Advanced Care Center, Sakurabashi Watanabe Hospital, Shinshu University Hospital, The University of Tokyo Hospital, Tohoku University Hospital, Tokyo Medical and Dental University Medical Hospital, Tokyo Women's Medical University Hospital, Tottori University Hospital, University of the Ryukyus Hospital.
Appendix 2 J-MACS investigators—Steering committee and operating committee members
Masatoshi Akiyama (Tohoku University), Hirokuni Arai (Tokyo Medical and Dental University), Masao Horie (Nipro Corporation), Atsushi Iguchi (Tohoku University), Tsutomu Imaizumi(International University of Health and Welfare), Takashi Inai (The Japan Federation of Medical Devices Associations), Hiroshi Ishikawa (The Japan Federation of Medical Devices Associations), Mitsuaki Isobe (Tokyo Medical and Dental University), Tohru Izumi (Niigata Minami Hospital), Takeo Katakura (Japan Medical Devices Manufacturers Association), Soichiro Kitamura (National Cerebral and Cardiovascular Center), Hiromi Kurosawa (Tokyo Women's Medical University), Shunei Kyo (Tokyo Metropolitan Geriatric Hospital), Hikaru Matsuda (Higashi Takarazuka Satoh Hospital), Goro Matsumiya (Osaka University), Takeshi Nakatani (Maki Hospital), Kan Nawata (The University of Tokyo), Takashi Nishimura (Tokyo Metropolitan Geriatric Hospital), Tomohiro Nishinaka (Tokyo Women's Medical University), Minoru Ono (The University of Tokyo), Hiroaki Oshiyama (Medical Technology Association of Japan), Yuki Ozawa (Terumo Corporation). Yoshikatsu Saiki (Tohoku University), Satoshi Saito (Tokyo Women's Medical University), Taichi Sakaguchi (Osaka University), Yasushi Sakata (Osaka University), Kazuhiro Sase (Juntendo University), Yoshiki Sawa (Osaka University), Koichi Tabayashi (Tohoku University), Yoshitaka Tada (Edwards Lifesciences Corporation), Yoshihisa Tanoue (Kyushu University), Koichi Toda (Osaka University), Ryuji Tominaga (Kyushu University), Masao Tozawa (Century Medical, Inc.), Keisuke Uratomi (Japan Federation of Medical Devices Associations), Hiroyuki Ushiyama (Sun Medical Technology Research Corp.), Takashi Yamane (National Institute of Advanced Industrial Science and Technology), Kenji Yamazaki (Tokyo Women's Medical University), Shunichi Yamazaki (Sun Medical Technology Research Corp.), Masanobu Yanase (National Cerebral and Cardiovascular Center).
Appendix 3 Cooperating academic societies
Japanese Society for Artificial Organs, Japanese Association for Clinical Ventricular Assist Systems, The Japanese Association for Thoracic Surgery, The Japanese Society for Cardiovascular Surgery, The Japanese Circulation Society, The Japanese Heart Failure Society, The Japanese College of Cardiology.
Appendix 4 Pharmaceuticals and Medical Device Agency (PMDA)
Tomiko Tawaragi, Mie Ikeda, Eri Kawabe, Ayumi Kishi, Emiko Kondo, Kazuhiro Matsui, Kaoru Misawa, Tomoko Momoki, Satoru Nakamura, Miho Sato, Mari Shirotani, Seiko Usami, Yoshiaki Uyama, Shinichi Watanabe.
Domestic and foreign trends in the prevalence of heart failure and the necessity of next-generation artificial hearts: a survey by the Working Group on Establishment of Assessment Guidelines for Next-generation Artificial Heart Systems.
U.S.-Japan Harmonization by Data (HBD)—initial experience with a post-marketing follow-up study of mechanical circulatory assist devices (J-MACS), and ongoing collaboration with academic research consortiums.
Pharmaceut Med Dev Regul Sci [in Japanese].2012; 43: 339-346
J-MACS Research Group: Post-approval study of a highly pulsed, low-shear-rate, continuous-flow, left ventricular assist device, EVAHEART: a Japanese multicenter study using J-MACS.
J-MACS Adverse Event and Adjudication committee. The first report on adverse event for mechanically circulatory support. December 2012. http://www.pmda.go.jp/files/000147200.pdf/. Accessed June 1, 2016.
31947-2&id=gr1.jpg){kind=link}
31947-2&id=gr2.jpg){kind=link}
31947-2&id=gr3.jpg){kind=link}
31947-2&id=gr4.jpg){kind=link}
31947-2&id=gr5.jpg){kind=link}
31947-2&id=gr6.jpg){kind=link}
31947-2&id=gr7.jpg){kind=link}