Hyperthermic preconditioning protects against spinal cord ischemic injury.

BACKGROUND: Paraplegia can result from operations requiring transient occlusion of the descending thoracic aorta. The present study tested whether inducing hyperthermia in rats before aortic ischemia would be neuroprotective. METHODS: Rats were randomly assigned to hyperthermic preconditioning (n = 27) or control (n = 32) groups. Eighteen hours before ischemia, the hyperthermic preconditioned rats were heated at 41 degrees C for 15 minutes. Ten minutes of spinal ischemia were produced by balloon occlusion of the thoracic aorta. Neurologic performance scores were evaluated daily to 7 days after ischemia. The lumbar region of the spinal cord was removed for histologic grading. RESULTS: The hyperthermic preconditioned animals had less permanent spinal cord injury compared with controls (29.6% versus 59.4%, p = 0.02), and the incidence of immediate paraplegia in the hyperthermic preconditioned group was significantly less than that in the control group (3.7% versus 28.1%, p = 0.03). Histologic scores correlated with the neurologic outcome at the time of sacrifice in rats with permanent spinal cord injury but not in those walking normally. CONCLUSIONS: We used a rat model of spinal cord ischemia and found that hyperthermic preconditioning before spinal cord ischemia resulted in improved clinical outcome.

Ischemic preconditioning protects against paraplegia after transient aortic occlusion in the rat.

BACKGROUND: Paraplegia can result from operations requiring transient occlusion of the thoracic aorta. A rat model of paraplegia with the characteristics of delayed paraplegia and transient ischemic dysfunction was developed to determine whether ischemic preconditioning (IPC) improved neurologic outcome. METHODS: Rats underwent balloon occlusion of the upper descending thoracic aorta. One group (2 minute IPC, n = 19) underwent 2 minutes of IPC and a second group (5 minute IPC, n = 19) had 5 minutes of IPC 48 hours before 10 minutes of occlusion. The control group (n = 31) had no IPC prior to 10 minutes of occlusion. RESULTS: Paraplegia occurred in 68% of the control animals (21 of 31 paraplegic: 6 delayed and 15 immediate paraplegia). Both the 2-minute IPC and 5-minute IPC groups had a decreased incidence of paraplegia when compared to controls (32%, p = 0.011 and 26%, p = 0.009, respectively). CONCLUSIONS: A rat model of spinal cord ischemia demonstrating both delayed paraplegia and transient ischemic dysfunction was characterized. Both 2-minute and 5-minute periods of IPC were found to protect against paraplegia.

IL-2 reduces graft-versus-host disease and preserves a graft-versus-leukemia effect by selectively inhibiting CD4+ T cell activity.

We have recently demonstrated, in a fully MHC-mismatched murine bone marrow transplantation model, that administration of a short course of high dose IL-2 markedly diminishes graft-vs-host disease (GVHD) without compromising alloengraftment or the graft-vs-leukemia (GVL) effect of allogeneic T cells. We have now evaluated the mechanism of the dissociation of GVL and GVHD observed in this model. We demonstrate that CD4+ T cells were required to produce severe, acute GVHD in the fully MHC-mismatched plus minor histocompatibility Ag-mismatched A/J-->B10 strain combination. The GVHD-producing activity of A/J CD4+ T cells administered without CD8+ T cells was inhibited by IL-2 treatment. In contrast, CD8+ T cells alone mediated the GVL effect observed in the EL4 leukemia/lymphoma model, and CD4+ cells did not contribute to this effect. This CD8-mediated GVL activity was not inhibited by IL-2 treatment. Because naive A/J CD8+ T cells administered without CD4+ T cells did not produce acute GVHD, we were unable to evaluate the effect of IL-2 in this model. However, when A/J donors were presensitized with B10 skin grafts, CD4-depleted A/J spleen cells were capable of causing acute GVHD in B10 recipients. This CD8-mediated GVHD was not inhibited by treatment with IL-2. However, IL-2 did partially inhibit the GVHD produced by nondepleted presensitized A/J spleen cells, probably due to selective inhibition of the function of presensitized A/J CD4+ T cells. The dissociation of GVHD and GVL against the EL4 leukemia/lymphoma in IL-2-treated mice can therefore be explained by selective inhibition by IL-2 of CD4 activity.

Mechanism of protection from graft-versus-host disease mortality by IL-2. III. Early reductions in donor T cell subsets and expansion of a CD3+CD4-CD8- cell population.

Reducing the graft-vs-host disease (GVHD)-promoting capacity of allogeneic T cells while maintaining alloengraftment and graft-vs-leukemia effects remains an important but elusive goal in clinical bone marrow transplantation (BMT). We have recently demonstrated that a short course of high dose IL-2 administered at the time of BMT has a powerful protective effect against GVHD mortality in mice. This short course of IL-2 is able to protect mice from both acute and chronic GVHD without sacrificing alloengraftment or graft-vs-leukemia effects of allogeneic T cells. Because the early administration of IL-2 seems to be crucial for this effect, we have studied the early lymphoid repopulation events after lethal irradiation and allogeneic BMT. These studies show that there are consistent delays in splenic repopulation by allogeneic cells after BMT in IL-2-treated animals compared with their untreated cohorts. Even greater percent reductions were seen in donor splenic T cell populations in the first few days after BMT in IL-2-treated animals. Splenic cells with the CD3+CD4-CD8- phenotype were increased in IL-2 treated animals at days 3 and 4 after BMT. This phenotype resembles that of bone marrow-derived cells which have been previously shown to inhibit GVHD, suggesting a possible mechanism for the protective effect of IL-2.

The mechanism of IL-2-mediated protection against GVHD in mice. II. Protection occurs independently of NK/LAK cells.

We have recently demonstrated that high-dose IL-2, when begun on the day of bone marrow transplantation, has a potent protective effect against graft-vs.-host disease mortality, especially when coadministered with T cell-depleted syngeneic bone marrow cells. Because several groups of investigators have demonstrated that lymphokine-activated killer cells can mediate GVHD protection, we hypothesized that the mechanism of protection by IL-2 administration might involve the in vivo activation of natural killer and/or LAK cells. In order to test this hypothesis, we evaluated the effect of IL-2 administration on the number of NK1+ cells and on NK-mediated cytotoxic activity in recipients of GVHD-producing inocula. Furthermore, we evaluated the effects on IL-2-induced GVHD protection of depleting NK cells and LAK precursor cells in vivo with mAb against NK1.1 or antiserum against asialo GM1. The results demonstrate that: (1) The number of NK1+ cells is not increased in spleens of IL-2-treated compared with control recipients of GVHD-producing inocula; (2) NK activity is not increased in IL-2-treated compared with control recipients of GVHD-producing inocula during or immediately following the period of IL-2 administration; (3) depletion of NK cells and LAK precursors from the donor and host influenced the time course of GVHD-related mortality in a complex fashion; and (4) IL-2-induced GVHD protection is largely independent of the activity of an NK or LAK cell population of donor or host origin. IL-2-induced GVHD protection therefore reflects primarily the activity of non-LAK protective cell populations, or it may be a direct inhibitory effect on responding donor cell populations as they encounter host antigen.

Mixed allogeneic chimeras prepared by a non-myeloablative regimen: requirement for chimerism to maintain tolerance.

We have recently described a non-myeloablative conditioning regimen permitting engraftment of allogeneic bone marrow in mice which involves administration of anti-CD4 (GK1.5) plus anti-CD8 (2.43) monoclonal antibodies in vivo, 3 Gy whole body irradiation, plus 7 Gy thymic irradiation. B10 (H-2b) mice prepared by this regimen and infused with unmanipulated B10.D2 (H-2d) bone marrow develop permanent mixed lymphohematopoietic chimerism and specific tolerance to donor skin grafts. We now demonstrate that mixed chimerism persists longer than 170 days in the lymphoid tissues including spleen, thymus and bone marrow of such animals, and that equivalent levels of donor chimerism are observed in both T and B cell compartments. In addition stable mixed chimeras were found to be unresponsive to host (B10) and donor (B10.D2) stimulator cells in mixed lymphocyte reaction and in cell mediated lympholysis assays, while responses to a third party (B10.BR, H-2k) were intact. Persistent chimerism was found to be necessary for the maintenance of skin graft tolerance in these animals, since in vivo depletion of donor cells by treatment with an anti-H-2d (34-2-12) monoclonal antibody resulted in the subsequent rejection of donor skin grafts. These studies demonstrate that mixed allogeneic chimeras produced using this regimen are specifically tolerant to donor in vitro and in vivo, and that persistence of donor chimerism is critical for the maintenance of tolerance.

Mechanism of the anti-GVHD effect of IL-2. I. Protective host-type cell populations are not induced by IL-2 treatment alone.

We have recently demonstrated that high-dose IL-2 administered for a short period (2.5 days) beginning on the day of bone marrow transplantation mediates a marked protective effect against GVHD in mice, while preserving the ability to achieve alloengraftment (1). This protective effect is augmented by administration of T cell-depleted (TCD) syngeneic marrow, and is dependent upon early administration of IL-2 (1). The graft-vs-tumor effect against the EL4 leukemia/lymphoma is not diminished in animals protected from GVHD by IL-2 (2). In an attempt to determine whether or not IL-2-activated host-type cells might be responsible for GVHD protection, we have now performed adoptive transfer studies. The results failed to provide evidence that treatment of lethally irradiated mice with IL-2 activates protective host-derived or syngeneic marrow-derived cell populations which can be adoptively transferred to lethally irradiated secondary recipients receiving allogeneic GVHD-producing inocula. Likewise, treatment of lethally irradiated mice with a complete 2.5-day course of IL-2 prior to administration of allogeneic inocula did not lead to GVHD protection. These results suggest that either IL-2 directly inhibits the GVH reactivity of allogeneic GVH-reactive cells, or that GVH reactivity is attenuated by IL-2 during the period of interaction of donor- and host-type cells.

Operative treatment of Crohn's ileocolitis complicated by ileosigmoid and ileovesical fistulae.

Ileovesical and ileosigmoid fistulae were found to coexist in 22 patients with Crohn's ileocolitis. Persistent or recurrent urinary tract infection was a complaint in all cases, and 11 patients reported pneumaturia and/or fecaluria. Thus, bladder involvement was either suspected or clinically apparent in each patient. The cystogram was the best confirmatory test for the ileovesical fistula (positive in 9 of 22 patients). The coexistence of the sigmoid fistula was best diagnosed on intestinal radiographs (positive in 9 of 22 patients); there were no clinical signs of its presence. The coexistence of ileosigmoid and ileovesical fistulae was the sole indication for operation in two patients. In all others, a combination of factors required surgical therapy. An ileocolonic resection with primary intestinal anastomosis was performed in 16 patients and exteriorization was performed in six patients. The sigmoid defect was closed primarily in 16 patients and required wedge resection in the other six patients. The bladder defect was sparingly excised and closed with absorbable sutures. All patients recuperated without anastomotic leaks, bladder leaks, or persistent cystitis. This experience indicates that coexisting ileosigmoid and ileovesical fistulae may add complexity to an ileocolonic resection for Crohn's disease, but is not a difficult management problem for the gastrointestinal surgeon.

Portal versus caval venous drainage of small bowel allografts: technical and metabolic consequences.

The potential metabolic and technical consequences of systemic (portacaval anastomosis [PC-A]) as opposed to portal venous drainage (portaportal anastomosis [PP-A]) of orthotopic small bowel isografts was evaluated in a rat model. Rats with portacaval (PC) shunts were studied for comparison. During the study period of 6 months, rats with small bowel grafts (PP-A or PC-A) gained weight at rates equal to that of normal age-matched rats (+40% of the preoperative weight) whereas rats with PC shunts lost 20% of their weight. At autopsy 6 months after operation, rats with PC shunts had significant liver atrophy (2.0% of total body weight) in comparison with rats with orthotopic isografts. Moderate liver atrophy was detected in rats with grafts and PC-A in comparison with those with PP-A (2.6% versus 2.8% of total body weight, statistically not significant). Serum ammonia levels were significantly elevated for PC shunts (560 +/- 148 micrograms/dl) and PC-A (140 +/- 22 micrograms/dl) when compared with PP-A (83 +/- 10 micrograms/dl). In terms of technical difficulties, both PC-A and PP-A could be achieved with the same success rate. Systemic venous drainage for small bowel grafts is followed by metabolic alterations that are similar, although much less pronounced, to those seen after a PC shunt. Thus our findings do not offer compelling reasons to prefer PP-A over PC-A. However, with longer follow-up and the use of hepatotoxic immunosuppressive drugs, these minimal alterations may progress and induce metabolic sequelae of clinical significance. Under these circumstances it would be advisable to use the physiologic portal drainage rather than systemic venous drainage in small bowel transplantation.

Portal versus systemic venous drainage for small-bowel allografts.

The effect of portal venous drainage (PV-D) on the survival of accessory small-bowel allografts was studied in a rat model. In the LBN-F1----LEW combination (rejection reaction only), grafts with caval venous drainage (CV-D) were rejected acutely (mean 11.8 days) and those with PV-D were rejected chronically (mean, 22.8 days). In the LEW----LBN-F1 combination, the survival of recipients subject perforce to a fatal graft-versus-host reaction was not influenced by the type of venous drainage used (CV-D: 14.1 days; PV-D: 14.8 days). These findings suggest that the delaying influence of the liver on the rejection process is not due to unspecific filtration of antigens but must involve their specific recognition or alteration. In the BN----LEW combination, in which rejection and the graft-versus-host reaction occur simultaneously, the type of venous drainage used did not influence graft survival (CV-D: 16.2 days; PV-D: 15.5 days), nor did it modify the rejection process. This indicates that the liver has a minor effect on the rejection process, sufficient to inhibit the rejection of semiallogeneic LBN-F1 grafts but insufficient to alter the rejection of allogeneic BN grafts by LEW recipients. With CV-D, LBN-F1 grafts were rejected as rapidly as were BN grafts. This unexpected finding may be explained by the fact that in the LBN-F1----LEW combination only rejection occurs, unimpeded by a graft-versus-host reaction, which is known to impair the immune system of the recipient; in the BN----LEW combination, however, the graft-versus-host reaction temporarily opposes the rejection reaction, thereby allowing prolonged graft survival.