Qiang Yin

Oregon Health and Science University, Portland, OR, United States

Are you Qiang Yin?

Claim your profile

Publications (20)54.07 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: To compare two techniques used to create a larger animal model of venous valve incompetence. To achieve vein dilatation as the primary cause of valve incompetence, common carotid jugular vein (JV) fistulas were created and optional filters were placed into the JV of sheep. Altogether, nine inferior vena cava filters were placed in three sheep in two stages. Six filters were placed caudal to the most caudal JV valve in three sheep and removed 6 weeks later. Then, three filters were placed across the most caudal valve in two sheep with competent valves and removed 3 weeks later. A common carotid artery-JV fistula was created in three sheep and followed-up for 1-3 weeks. Ascending and descending venograms were obtained to determine the JV sizes and function of their valves. The JVs removed at necropsy were studied with venoscopy. Only one of the six JVs with filters caudal to the most caudal valve had incompetent valves after filter removal at 6 weeks. In addition, only one of three JVs with the filter across the valve had incompetent valves after filter removal at 3 weeks. At 1-3-week follow-up of the group with common carotid artery-JV fistula, all three JVs had incompetent valves in the cephalad vein portion, but only one JV had an incompetent valve in its caudal portion. At venoscopy, the incompetent valves showed various degrees of damage ranging from shortening to the destruction of valve leaflets. Dilation of the valve annulus with a removable vena cava filter failed to produce valve incompetence. The promising results with the common carotid artery-JV fistula justify further detailed research.
    Journal of Vascular and Interventional Radiology 10/2007; 18(9):1169-75. · 2.00 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Limited experience with bioprosthetic venous valve percutaneously inserted into femoral veins in 15 patients has been promising in short-term results only to show disappointing long-term results. Percutaneous autogenous venous valve (PAVV) transplantation was explored in an ovine model as a possible alternative treatment. PAVV consisted of a vein segment containing a valve that was attached to a stent template. The stent templates (n = 9) were designed and hand made in our research laboratory. They consist of two stainless steel square stents 13 or 15 mm in diameter to fit the ovine jugular veins (JV), which ranges from 10 to 15 mm in diameter. A valve-containing segment of JV was harvested and attached with sutures and barbs inside the stent template (n = 9). The valve devices were then manually folded and front loaded inside the 4 cm chamber of the 13F delivery sheath and delivered into the contralateral JV by femoral vein approach. Transplanted PAVVs were studied by immediate and 3 months venograms. Animals were euthanized at 3 months, and jugular veins harvested to perform angioscopic evaluations in vitro. PAVV transplantation was successful in all nine animals. Good valve function with no reflux was observed on immediate and 3 months venograms in eight valves. The transplanted maximal JV diameter ranged from 10.2 mm to 15.4 mm (mean 13.1 +/- 1.5 mm). Venoscopic examination revealed intact, flexible, nonthickened valve leaflets in eight specimens. One PAVV exhibited normal function of one leaflet only; the other cusp was accidentally cut during the transplantation procedure. All transplanted autologous valves were free of thrombus and incorporated into the vein wall of the host vessel. This study demonstrated that autogenous valve transplants remained patent and competent without long-term anticoagulation for up to 3 months. The percutaneous autogenous venous valve may provide in future minimally invasive treatment for patients with chronic deep venous insufficiency, but long-term studies need to be done to document its continued patency and function.
    Journal of Vascular Surgery 09/2007; 46(2):338-45. · 2.88 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To explore the feasibility and efficacy of residual aneurysmal sac (RAS) embolization with the expandable hydrogel embolic device (EHED) in prevention of endoleaks in a surgically created and endoluminally treated abdominal aortic aneurysm (AAA). In eight dogs, an AAA was created by means of side-to-side anastomosis between the infrarenal abdominal aorta and inferior vena cava (IVC) with ligation of the IVC above and below the anastomotic end, followed by deployment of an endograft with holes. The RAS was then embolized with the EHED. One animal was killed immediately after RAS embolization and one animal died 12 hr after the procedure. Follow-up aortograms were obtained in six animals after 1 day (1 animal), 2 weeks and 6 months (1 animal), and 8 weeks (4 animals). Four animals had no endoleaks on the follow-up aortograms. The remaining two animals with incomplete RAS embolization had moderate type III endoleaks. Type I or II endoleaks were not seen in any animals. Complications included RAS wall penetration by the devices with platinum wires in two animals (nos. 1 and 2), device migration into an aortic circulation through the endograft holes in two animals (nos. 2 and 3) or through distal interstices between the aortic wall and endograft in one animal (no. 8), aortic occlusion in three animals (nos. 3, 7, and 8), and RAS rupture in one animal (no. 7). Histologic examination showed expanded hydrogels occupying the RAS with associated mature or immature organized thrombus, fibrinous thrombus, or degenerate blood cells. RAS embolization was feasible with the EHED, although additional modifications to the device are required to avoid complications. Angiographic and histologic results suggested that RAS embolization with the EHED may help in the prevention of endoleaks.
    CardioVascular and Interventional Radiology 05/2005; 28(4):459-66. · 2.14 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To percutaneously create an improved abdominal aortic aneurysm model of endoleak after endograft placement and to explore efficacy of small intestinal submucosal embolization of the residual aneurysmal sac for prevention of endoleaks. Abdominal aortic aneurysm was created transluminally by over-dilation of a Palmaz stent in 12 sheep. Approximately 20% undersized endografts suspended between two stent-graft adapters were used to bridge the aneurysm in a manner that two lumbar pairs remained patent within the residual aneurysm sac. Size of the residual aneurysm sac was increased by placement of an undersized stent-graft consisting of damaged lyophilized small intestinal submucosal sheets sandwiched between two Zilver stents. In six sheep, residual aneurysm sacs were embolized by combining small intestinal submucosal sponge and small intestinal submucosal sheet pieces. The other six sheep served as the control group. Angiography performed immediately after the procedure was compared with follow-up angiography before the animals were killed at 1, 3, and 7 months. Gross and histologic examinations were also obtained. Aortic ruptures (n = 3) and dissections (n = 2) during aneurysm creation responded well to endograft placement. Eleven endografts were placed successfully, one was misplaced. The mean diameter of aneurysmal sac was 16 mm in the study and 15.2 mm in the control group. In the study group, in four sheep, the sac and seven pairs of lumbar arteries were occluded by embolization and remained obstructed by organized thrombus during the entire study. There were no type II endoleaks. Four type III new endoleaks developed without antegrade filling of lumbar arteries. In the control group, five animals had type I and II endoleaks at the initial studies. Only one sheep exhibited completely organized thrombosis of the aneurysmal sac and without endoleaks. In the other four sheep with partially organized sac thrombosis, endoleaks were unchanged. One type III endoleak occurred in this group. The combination of small intestinal submucosal sponge and small intestinal submucosal sheet pieces is a promising embolic material for occlusion of the residual sac after endovascular abdominal aortic aneurysm repair and for prevention of type II endoleaks.
    Journal of Vascular and Interventional Radiology 02/2004; 15(1 Pt 1):69-83. · 2.00 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this study was to create an animal model of endoleak after stent-graft placement for abdominal aortic aneurysm (AAA) in which a large aneurysmal sac would be preserved for the testing of techniques for its percutaneous occlusion. Infrarenal AAAs were created in nine dogs by anastomosis of an isolated segment of the inferior vena cava to the right side of the abdominal aorta in combination with a large anterior patch from the external jugular vein. One hour later, animals underwent percutaneous implantation of polytetrafluoroethylene-covered Z stent endografts with three 3-mm-diameter holes through the fabric. Aortograms were obtained before and after surgery, after endograft placement, and at the time of animal sacrifice at 1 week or 1, 2, 3, or 6 months. Pressures within the aorta and the aneurysm sac were recorded before animal sacrifice. Gross and histologic evaluations of the specimens were then carried out. Immediately after endograft placement, all nine animals had artificial type III endoleaks with angiographic filling of lumbar arteries and veins. One animal died of surgical complications within 2 days of surgery and is not included in our data analysis. One aneurysm ruptured at 1 week. At completion of the study, six endografts were patent and two were occluded. The aneurysm sac had enlarged by approximately 50% in seven animals. At follow-up, type I endoleak was present in three animals, type II endoleak was present in three, and the artificial type III endoleak was present in all six animals with patent endografts. The pressure differential between aorta and aneurysm sac was 36 mm Hg, with a mean aortic pressure of 87 mm Hg +/- 13.3 and a mean aneurysmal sac pressure of 51 mm Hg +/- 28.1. The aneurysmal sac exhibited early thrombus formation at 1 week, which progressed to complete thrombosis in 1-6 months. The model is technically feasible but would be useful in testing occlusive techniques for residual aneurysm sacs only in the acute phase after endograft placement. It would be not reliable for chronic evaluation because of rapidly progressive thrombosis in most aneurysm sacs and occasional complete thrombosis of the AAA and endograft.
    Journal of Vascular and Interventional Radiology 11/2003; 14(10):1303-10. · 2.00 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Cytomegalovirus (CMV) infections have been shown to dramatically affect solid organ transplant graft survival in both human and animal models. Recently, it was demonstrated that rat CMV (RCMV) infection accelerates the development of transplant vascular sclerosis (TVS) in both rat heart and small bowel graft transplants. However, the mechanisms involved in this process are still unclear. In the present study, we determined the kinetics of RCMV-accelerated TVS in a rat heart transplant model. Acute RCMV infection enhances the development of TVS in rat heart allografts, and this process is initiated between 21 and 24 days posttransplantation. The virus is consistently detected in the heart grafts from day 7 until day 35 posttransplantation but is rarely found at the time of graft rejection (day 45 posttransplantation). Grafts from RCMV-infected recipients had upregulation of chemokine expression compared to uninfected controls, and the timing of this increased expression paralleled that of RCMV-accelerated neointimal formation. In addition, graft vessels from RCMV-infected grafts demonstrate the increased infiltration of T cells and macrophages during periods of highest chemokine expression. These results suggest that CMV-induced acceleration of TVS involves the increased graft vascular infiltration of inflammatory cells through enhanced chemokine expression.
    Journal of Virology 03/2003; 77(3):2182-94. · 5.08 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To compare the biocompatibility and performance of nitinol endografts covered externally or internally with small intestinal submucosa (SIS) with bare nitinol stents in medium-sized arteries. Eighteen nitinol Zilver stents were used: six externally SIS-covered endografts (ECEs), six internally SIS-covered endografts (ICEs), and six bare stents (BSs). Devices were implanted in the balloon-injured external iliac arteries (EIAs) of nine female sheep via carotid approach. Arteriograms were obtained before and after implantation and before animal sacrifice at 1, 3, and 6 months. Histologic studies of explanted specimens were performed. Implantation of all BSs, ECEs, and ICEs was successful, but slight luminal narrowing of 19% +/- 5.3% (range, 12%-28%) was seen in ICEs on postimplantation angiograms. At sacrifice, all six BSs and ECEs were patent, with BSs showing a mean angiographic luminal narrowing of 8.4% +/- 7.2% (range, 0%-18%) and ECEs showing a mean angiographic luminal narrowing of 16% +/- 7.5% (range, 6.5%-26%) as a result of neointimal hyperplasia. Four ICEs showed luminal narrowing of 21% +/- 17% (range, 0%-35%) as a result of marked neointimal hyperplasia and two were occluded, one at 3 months and the other at 6 months. ECEs and BSs showed approximately 80% lumen endothelialization at 1 month, which increased to 100% at 3 and 6 months. ICEs did not show complete endothelialization. The BS had the least vessel wall reaction. ECEs exhibited early endothelialization with early mild to moderate wall reaction decreasing at the late study stages. ICEs showed extensive wall reaction, possibly as a result of technical problems with SIS attachment.
    Journal of Vascular and Interventional Radiology 06/2002; 13(5):489-98. · 2.00 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The primary cause for late failure of vascularized allografts is chronic rejection (CR) characterized by transplant vascular sclerosis (TVS). Cytomegalovirus (CMV) infection accelerates TVS and CR by unclear mechanisms involving direct effects of CMV, indirect effects of the recipient's immune response to CMV, or interactions between CMV and the recipient's alloreactivity. This study examined the role of CMV and the alloreactive response in the development of TVS using bone marrow chimerism (BMC) in rat small bowel (SB) and heart transplantation models. Fisher 344 (F344) rat heart or SB grafts were transplanted into F344/Lewis bone marrow chimera. F344 heart or SB grafts transplanted into Lewis recipients (low-dose cyclosporine) were positive controls for the development of TVS. Lewis heart or SB grafts transplanted into Lewis recipients (+/-cyclosporine) were transplantation controls. The effect of rat CMV (RCMV) (5x105 plaque-forming units) on TVS (neointimal index, NI) and graft survival was studied in these groups. RCMV infection was assessed by serologic analysis and quantitative polymerase chain reaction techniques (TaqMan). RCMV infection accelerated the time to graft CR (SB 70-38 days; hearts 90-45 days) and increased the severity of TVS in both the SB allografts (day 38, NI=27 vs. 52) and the heart allografts (day 45, NI=43 vs. 83). Grafts from CMV-infected syngeneic recipients failed to develop TVS and CR. Donor-specific tolerance induced by BMC prevented allograft TVS and CR in both transplant models. In contrast to naïve Lewis recipients, RMCV infection failed to cause allograft TVS and CR in bone marrow (BM) chimeras. The events in CMV-induced acceleration of TVS involve a crucial interplay between CMV infection and the recipient's alloreactive immune response.
    Transplantation 04/2002; 73(5):679-88. · 3.78 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Chronic rejection (CR) and transplant vascular sclerosis (TVS) cause the majority of graft failures in cardiac transplantation. Hyperhomocyst(e)inemia [hH(e)] is associated with human TVS without a proven causal relationship. This study investigated the effect of hH(e) on graft survival and TVS in allogeneic and syngeneic rat cardiac transplants. Lewis recipients of heterotopic F344 heart allografts, received normal or hH(e)-inducing (folate, methionine) diets [controls: syngeneic transplanted [+/- hH(e), + CsA] and nontransplanted rats [+/- hH(e), +/- CsA]]. Serial plasma homocyst(e)ine [H(e)] levels were measured. TVS was assessed in clinically rejected grafts and a subset of pre-rejection normal diet allografts (day 64) (neointimal index, NI). The hH(e) diet elevated plasma H(e) levels. When compared with normal diet controls (n = 9), hH(e) diet allografts (n = 9) had decreased time to onset of CR (40 +/- 9 vs. 72 +/- 10d, p = 0.02), and graft failure (64 +/- 10 vs. 107 +/- 12d, p = 0.009). hH(e) diet allografts at rejection (n = 9, 64d) had more severe TVS (NI = 68 +/- 2) than both time-matched normal diet allografts (NI = 49 +/- 6, n = 8, 64d, p <0.001) and normal diet allografts at rejection (NI = 58 +/- 5, n = 9, 107d, p = 0.007). hH(e) induced TVS in syngeneic grafts (NI=50 +/- 3, n = 10 vs. NI = 5 +/- 3, n = 10, 130d, p <0.001). hH(e) accelerated rejection and increased the severity of TVS in allogeneic cardiac transplants, and induced TVS in syngeneic cardiac transplants.
    American Journal of Transplantation 03/2002; 2(3):244-51. · 6.19 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate a new, low profile, home-made, bifurcated drum occluder endograft (BDOEG), designed for percutaneous, transcatheter treatment of abdominal aortic aneurysm (AAA). AAA was created in 10 dogs with over-dilated Palmaz stents. To prevent back filling, the lumbar arteries, inferior mesenteric artery, and common internal iliac arteries were embolized. The BDOEG was constructed of a drum occluder device and two PTFE endografts. The drum device consisted of a modified Z stent with Dacron stretched across and held within the ends of the stent, each with two 8 x 6-mm slits through which PTFE endografts were delivered. The PTFE endografts were 8 mm in diameter and 9.5 cm in length. Preloaded, the BDOEG was delivered through a 10-F sheath from both femoral arteries in a three-step procedure. All 10 animals were treated with BDOEG. Aortography was performed immediately, 6 weeks, and 12 weeks after stent-graft placement. Five animals were killed at 6 weeks and five were killed at 3 months. Gross and histologic evaluation was performed. The infrarenal aortic diameters and both external iliac arteries ranged from 8.0 mm to 10.3 mm (mean, 9.4 mm +/- 0.6) and from 5.2 mm to 6.8 mm (mean, 5.8 mm +/- 0.5), respectively. Creation of the AAA was successful in all 10 dogs. AAA diameters ranged from 13.7 mm to 15.9 mm (mean, 14.9 mm +/- 0.7). Complete exclusion of the AAA was achieved immediately after BDOEG placement and aneurysms remained excluded without perigraft leak to the time of killing in all 10 animals. There was a high incidence of aortoiliac limb occlusion. Occlusion of 12 aortoiliac limbs (60%) caused by intimal hyperplasia at the distal end of the endografts in iliac arteries developed in nine animals (90%). In six animals (60%), one limb occluded and, in three animals (30%), there was occlusion of both limbs. This study suggests a new approach for treatment of AAA. BDOEG use reduces sheath size for endograft delivery and may eliminate the need for a surgical cut down on femoral arteries. Tapering of the iliac ends of endografts to the size of the artery will be needed to prevent distal intimal hyperplasia.
    Journal of Vascular and Interventional Radiology 04/2001; 12(3):359-64. · 2.00 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate efficacy of small intestinal submucosa (SIS) Sandwich endografts for the treatment of acute rupture of abdominal aortic aneurysms (AAA) and to explore the short-term reaction of the aorta to this material. In eight adult sheep, an infrarenal AAA was created transluminally by dilation of a short Palmaz stent. In six sheep, the aneurysm was then ruptured by overdilation of the stent with a large angioplasty balloon. Two sheep with AAAs that were not ruptured served as controls. A SIS Sandwich endograft, consisting of a Z stent frame with 5 bodies and covered inside and out with SIS, was used to exclude the ruptured and non-ruptured AAAs. Follow-up aortography was done immediately after the procedure and before sacrifice at 4, 8, or 12 weeks. Autopsy and histologic studies followed. Endograft placement was successful in all eight sheep. Both ruptured and non-ruptured AAAs were successfully excluded. Three animals with AAA rupture developed hind leg paralysis due to compromise of the arterial supply to the lower spinal cord and were sacrificed 1 day after the procedure. In five animals, three with rupture and two controls, follow-up aortograms revealed no aortic stenoses and no perigraft leaks. Gross and histologic studies revealed incorporation of the endografts into the aortic wall with replacement of SIS by dense neointima that was completely endothelialized in areas where the endograft was in direct contact with the aortic wall. In central portions of the endograft, in contact with the thrombosed aneurysm, endothelialization was incomplete even at 12 weeks. The SIS Sandwich endografts effectively excluded simple AAAs and ruptured AAAs. They were rapidly incorporated into the aortic wall. A detailed long-term study is warranted.
    CardioVascular and Interventional Radiology 01/2001; 24(2):99-105. · 2.14 Impact Factor
  • Transplantation Proceedings 01/2001; 33(1-2):1822-3. · 0.95 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The major impediment to success in solid organ transplantation is chronic rejection (CR). The characteristic lesion of CR is transplant vascular sclerosis (TVS). Although the mechanism of TVS is thought to have an immunologic basis, in humans immunosuppression does not prevent or reverse it. One possible therapy to prevent TVS is induction of donor-specific tolerance. Bone marrow chimerism has been successful in inducing tolerance in acute and chronic rejection heart and kidney transplant models. The highly immunogenic small bowel (SB) allograft provides a rigorous test of the efficacy of this tolerance regimen. We examined whether induction of tolerance by bone marrow chimerism could prevent TVS in a model of Fisher 344 (F344) to Lewis (LEW) rat SB transplantation. Bone marrow chimeras (BMC) were created by transplantation of T-cell-depleted F344 bone marrow into irradiated LEW rats. Chimerism was assessed by flow cytometric method. F344 SB, heterotopically transplanted into the chimeras, was clinically and histologically assessed for CR. F344 SB grafts, transplanted into cyclosporine-A-treated LEW recipients, served as control grafts for CR. Cyclosporine-A-treated LEW rats chronically rejected F344 SB grafts. By contrast, the BMC group demonstrated tolerance and had long-term SB graft survival (>120 days) without TVS. The BMC demonstrated immunocompetence by prompt rejection of third party ACI (RT1av1) SB allografts. Bone marrow chimerism prevents chronic graft failure secondary to TVS in a model of chronic SB rejection. TVS fails to develop when tolerance is established, suggesting that the mechanisms involved in TVS are, in part, immunologically mediated.
    Transplantation 04/2000; 69(7):1295-303. · 3.78 Impact Factor
  • Journal of The American College of Surgeons - J AMER COLL SURGEONS. 01/2000; 191(4).
  • Journal of The American College of Surgeons - J AMER COLL SURGEONS. 01/2000; 191(4).
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The major impediment to long-term success in solid organ transplantation is the development of chronic rejection (CR). The vascular lesion of CR, transplant vascular sclerosis (TVS) is characterized by neointimal smooth muscle cell proliferation, and is driven by both immune- and nonimmune-mediated mechanisms. Although the features of chronic heart and kidney allograft rejection have been well characterized, the more immunogenic small bowel allograft has not received similar study. F344 small bowel (SB) was transplanted heterotopically into Lewis recipients that were treated with low-dose Cyclosporine A for 15 days. Lewis recipients of F344 or Lewis SB grafts without immunosuppression, served as controls. Grafts were assessed histologically when recipients showed clinical signs of rejection or at predetermined time points. The immunological components involved in the chronic rejection process were evaluated by immunohistochemical staining. All SB allografts (100%) developed histologic evidence of CR Cyclosporine A. TVS was seen in 36 of the 46 (78%) of these allografts. The median time to develop TVS was 45 days. Immunohistochemical staining of chronically rejected grafts showed infiltration predominantly by CD4+ cells and macrophages, uniform up-regulation of class II MHC molecule expression, moderate to intense ICAM-1 staining in grafts harvested at postoperative day 45, and uniform neointimal cell staining for smooth muscle cell alpha-actin in the TVS lesions. This F344 to Lewis SB transplant model is a useful model that reproduces significant features of CR. The highly immunogenic nature of the SB allografts allows this model to serve as a stringent test for protocols designed to prevent CR.
    Transplantation 10/1999; 68(6):766-79. · 3.78 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To explore feasibility of twin-tube endografts (TTEGs) for treatment of abdominal aortic aneurysm (AAA). AAAs were created in six dogs by overdilation of Palmaz stents. TTEGs made of seven 9.5-mm-diameter Gianturco Rösch Z (GRZ) stents and covered with PTFE were deployed simultaneously from both femoral arteries through 10-F sheaths to exclude the AAA. Follow-up aortography was performed immediately and at 2, 4, and 8 weeks. Two dogs were killed at 4 weeks and four were killed at 8 weeks. Gross and histologic examinations were performed. AAA creation and TTEG placement were successful in all six dogs. TTEGs completely excluded AAAs in five dogs. Minor temporary perigraft leak was seen in one dog immediately after stent placement. Four dogs had TTEGs patent to termination at 8 weeks. In two dogs with oversized TTEGs, one of the twin endografts occluded at 2 weeks. At autopsy, TTEGs were formed in semicircular shape and histology did not show any significant inflammatory reaction. TTEGs show promise for treatment of AAA, with the potential of reducing the introductory sheath size. Further experimental studies, particularly with TTEGs extended into iliac arteries, are warranted.
    Journal of Vascular and Interventional Radiology 10/1999; 10(8):1092-8. · 2.00 Impact Factor
  • Transplantation 04/1999; 67(7):S250. · 3.78 Impact Factor
  • Transplantation 01/1999; 68(6):766-779. · 3.78 Impact Factor
  • Transplantation 05/1998; 65(5):115. · 3.78 Impact Factor