Salim Aziz

Prospective Trial of Catheter Irrigation and Muscle Flaps for Sternal Wound Infection

BACKGROUND Sternal wound infection is a relatively rare but potentially devastating complication of open heart operations. The most common treatments after debridement are rewiring with antibiotic irrigation and muscle flaps. Here we present the results of a prospective trial to determine the appropriate roles of closed-chest catheter irrigation and muscle flap closure for sternotomy infection and to assess the effect of internal mammary artery bypass grafting on the outcome of each treatment modality. METHODS Between 1990 and 1994, 5,658 sternotomies were performed at the University of Washington Medical Center. Sternal dehiscence occurred in 43 patients, 25 of whom had infection (overall incidence, 0.44%). Because of the infrequency of this complication, a prospective, randomized trial was developed in which the initial approach to sternal dehiscence was rewiring and catheter irrigation. Muscle flaps were used as the primary treatment if the sternum could not be restabilized or as secondary treatment if catheter irrigation failed. Wound resolution, length of hospital stay, and complications were evaluated. RESULTS Sterile dehiscences were successfully closed with irrigation in 17 of 18 patients; the other patient required flap closure. Of the 25 patients with infection, 19 had irrigation and 6, closure with flaps primarily. In the group of infected patients, 17 of the 19 who received irrigation also had internal mammary artery bypass grafting. Irrigation failed in 15 (88.2%) of these 17 patients, and salvage was accomplished with muscle flap closure. All 6 patients with infection who were closed primarily with muscle flaps had a successful outcome. Hospitalization averaged 10.2 days when muscle flaps were used primarily and 14.3 additional days for unsuccessful irrigation. When irrigation was successful, the hospital stay averaged 11.2 days. CONCLUSIONS Catheter irrigation should be reserved for patients without infection or patients with infection but without internal mammary artery bypass grafts in whom dehiscence occurs less than 1 month after sternotomy. All others should have closure with muscle flaps.

Endothelial loss and regeneration in a model of transplant arteriosclerosis

Early endothelial injury may play a role in the development of transplant arteriosclerosis. The present study documents early endothelial changes using a rat aortic graft model. Abdominal aortic allografts from PVG rats were orthotopically transplanted to DA rats. Controls were DA to DA transplants. Endothelial cell (EC) injury, regeneration, and leukocyte infiltration in the intima were evaluated using scanning electron microscopy and histological and immunocytochemical techniques. Nontransplanted aortic segments showed partial loss of ECs after 1 or 2 hr of preservation. Control isografts demonstrated extensive EC denudation and neutrophil adherence to residual ECs at 1 day post-transplantation. After 3 days, isografts showed continued regeneration of ECs in the central area and ingrowth of endothelium from both clamped sites in the recipient aorta. Reendothelialization was complete by day 14. Allografts showed similar findings to isografts up to day 3. In contrast to isografts, however, there was a secondary EC loss beginning at day 7. Monocytes/macrophages and T cells were noted to be adherent to residual ECs in 7- and 14-day allografts. At 20 days, ECs were absent from the luminal surface in the center of allografts. Endothelium did extend from clamped sites toward the midgraft region as in isografts. By 60 days allografts were completely reendothelialized. These results demonstrate that in both isografts and allografts there is initial EC loss due to mechanical trauma and ischemia/reperfusion injury, followed by partial reendothelialization. This latter process continues unabated in isografts, whereas in allografts the secondary EC loss occurs due to an allogenic response. This is followed by complete reendothelialization that occurs during the concurrent development of significant intimal hyperplasia.

Oral presentation of posttransplantation lymphoproliferative disorders. An unusual manifestation.

Cyclosporine, an immunosuppressive agent widely used in organ transplantation, has several undesirable side effects, including gingival hyperplasia, which occurs in up to 70% of patients. Another complication associated with use of cyclosporine and other immunosuppressants is an increased incidence of malignancies. Long-term use of cyclosporine also is associated with a spectrum of hyperproliferative disorders ranging from reactive lymphoid hyperplasia to aggressive malignant lymphomas. While cyclosporine-related lymphoproliferative disorders have been widely reported, they have not been described in the oral cavity as the first manifestation of this disease. We report on two cardiac transplantation patients with a history of cyclosporine use who presented initially with oral symptoms of lymphoproliferative disorder. Both had erythematous to cyanotic and hyperplastic gingiva. On gingivectomy, the fixed tissue was soft, glistening, and tan colored, in contrast to the usual firm, white, cyclosporine-associated, benign gingival fibrous hyperplasia. Histologically, a dense, diffuse infiltrate of lymphoplasmacytoid cells with vesicular nuclei, prominent nucleoli, a moderate amount of cytoplasm, and high mitotic activity was observed. Immunocytochemical studies confirmed that the cells were monoclonal for lambda light chains in one patient and kappa light chains in the other. The cells from one patient were positive for CD45, while both patients were negative for CD20 and all nonhematopoietic antigens tested. Both tissues were strongly positive for Epstein-Barr virus. Morphology and immunocytochemistry findings are consistent with a posttransplant lymphoproliferative disorder. These are the first two reported cases of cyclosporine-associated posttransplant lymphoproliferative disorders presenting as gingival hyperplasia.

Blood ionized magnesium concentrations during cardiopulmonary bypass and their correlation with other circulating cations

Abstract Background and aim: The recent introduction of new measurement technology (using ion specific electrodes) makes intraoperative evaluation of blood ionized magnesium (Mg2+, or iMg)–the bioactive fraction of circulating magnesium–possible. The goals of this study were: (1) to examine the longitudinal pattern(s) of change in blood iMg during cardiopulmonary bypass (CPB); and (2) to determine the relationship of iMg to Ca2+ (iCa), K, pH, Na, and hematocrit (Hct) during CPB. Methods: Blood was collected serially before, during, and after CPB on 30 patients undergoing elective coronary artery bypass graft procedures and the iMg was measured with an AVL Scientific Corp., model 988–4 instrument. Results: Overall, 73% of iMg results were abnormally low, 50% during CPB. Some cases had both hypo‐ and hyperionized magnesemic episodes. There were low iCa during CPB in 97% of cases. Using Spearmans rank order correlations and p < 0.05, iMg and K were directly correlated before, during, and after bypass, suggesting their parallel movement between tissue and blood. iMg and iCa were directly correlated before, and inversely correlated after, CPB, but unassociated during bypass. iMg and Na were inversely correlated after bypass in all cases. iMg was inversely correlated to pH and positively correlated to Hct during CPB only, and only in patients with concurrent association of iMg and iCa. Conclusions: Blood iMg depletion occurs frequently in CPB patients. iMg changes are not readily predictable. The association of intraoperative iMg depletion with postsurgical atrial fibrillation–reported to have a hypomagnesemic connection–should be investigated.

Clinical assessment of intraarterial blood gas monitor accuracy

The accuracy of intraarterial blood gas monitoring (IABGM) devices is challenging to assess under routine clinical conditions. When comparing discrete measurements by blood gas analyzer (BGA) to IABGM values, it is important that the BGA determinations (reference method) be as accurate as possible. In vitro decay of gas tensions caused by delay in BGA analysis is particularly problematic for specimens with high arterial oxygen tension (PaO2) values. Clinical instability of blood gases in the acutely ill patient may cause disagreement between BGA and IABGM values because of IABGM response time lag, particularly in the measurement of arterial blood carbon dioxide tension (PaCO2). We recommend that clinical assessments of IABGM accuracy by comparison with BGA use multiple bedside BGA instruments, and that blood sampling only occur during periods when IABGM values appear stable.