High levels of CD8-positive lymphocytes expressing CD45R0, granzyme B, and Ki-67 in lymph nodes of HIV-infected individuals are not associated with increased mortality.

Lymph nodes constitute the major site of HIV replication and of immunological response to HIV. To study the role of cytotoxic and mitotic active CD8(+) lymphocytes in lymph nodes during HIV infection we examined 28 formalin-fixed, paraffin-embedded lymph nodes sampled from 1984 to 1986 from 21 HIV-seropositive patients and seven HIV-negative patients. Eleven of the HIV-positive patients died within 78 months of biopsy time and 10 patients were alive on July 1, 1998. Double immunohistochemical staining procedures were developed to identify CD8(+) cells expressing CD45R0, granzyme B, and Ki-67. A stereological method was used to count the different cell types in the lymph nodes. There were no significant differences in the total cell (nucleated) and CD3(+) cell concentrations between the three groups. However, there were significantly higher concentrations of CD3(+)CD8(+), CD8(+)CD45R0(+), and CD8(+)Ki-67(+) lymphocytes in the HIV patients compared with the control group. Furthermore, there was a tendency for the HIV-deceased group to have lower levels of CD8(+)granzyme B(+) and CD8(+)Ki-67(+) lymphocyte concentrations compared with the HIV-alive group. Three HIV patients, who progressed to death within 49 months of biopsy time, were among the patients with the lowest concentrations of CD8(+)granzyme B(+) and CD8(+)Ki-67(+) lymphocytes. This finding allowed us to conclude that CD8(+) lymphocytes expressing high levels of CD45R0, granzyme B, and Ki-67 in lymph nodes of HIV patients are not related to increased mortality, whereas low concentrations of CD8(+) granzyme B(+) and CD8(+)Ki-67(+) lymphocytes may be associated with poor prognosis.

Granulocyte colony-stimulating factor increases CD4+ T cell counts of human immunodeficiency virus-infected patients receiving stable, highly active antiretroviral therapy: results from a randomized, placebo-controlled trial.

Thirty human immunodeficiency virus (HIV)-infected patients with CD4+ T cell counts <350 cells/mm3 who had received stable, highly active antiretroviral therapy (HAART) for at least 24 weeks were randomized to receive either placebo or granulocyte colony-stimulating factor (G-CSF; 0.3 mg/mL 3 times a week) for 12 weeks. Blood samples were collected at specified time points. G-CSF treatment enhanced the total lymphocyte count (P=.002) and increased CD3+ (P=.005), CD4+ (P=.03), and CD8+ (P=.004) T cell counts as well as numbers of CD3-CD16+CD56+ NK cells (P=.001). The increases in CD4+ and CD8+ cell counts resulted from increases in CD45RO+ memory T cells and cells expressing the CD38 activation marker. Lymphocyte proliferative responses to phytohemagglutinin and Candida antigen decreased, whereas NK cell activity and plasma HIV RNA did not change during G-CSF treatment. After 24 weeks, all immune parameters had returned to baseline values. This study suggests that G-CSF treatment of HIV-infected patients receiving stable HAART increases the concentration of CD4+, CD8+, and NK cells without inducing changes in the virus load.

Comparison of cold versus warm cardioplegia. Crystalloid antegrade or retrograde blood?

BACKGROUND: To evaluate the efficacy of warm versus cold and antegrade versus retrograde cardioplegia, 163 patients were randomized in sequence in three groups and underwent isolated coronary artery bypasses (mean, 4 grafts/patients) alternating in sequence. METHODS AND RESULTS: The demographic profiles were identical. Cold crystalloid cardioplegia (group 1) was delivered through the aortic root intermittently. The warm retrograde (group 2) was started antegrade with high potassium solution until the heart stopped. It was continued with retrograde coronary sinus perfusion, 100 mL/min of warm blood. Cold retrograde blood cardioplegia (4:1) (group 3) was started antegrade and continued retrograde through the coronary sinus. The heart temperature was kept at 10 degrees to 15 degrees C. Patients were evaluated intraoperatively and postoperatively for (1) supraventricular tachycardia, (2) ECG changes, (3) lactate dehydrogenase and total CPK and isoenzyme (MB) studies, and (4) hemodynamic studies in the intensive care unit. Warm retrograde and cold retrograde patients had sampling of the ascending aorta (antegrade) and the coronary sinus (retrograde) measuring pH, A-VO2 differences, and CK enzyme leak. RESULTS: The incidence of supraventricular tachycardia was 29% in group 2, 22% in group 1, 18% in group 3; not significant (NS). CPK isoenzyme MB fraction showed identical levels (NS). The warm heart consumed 1.3 to 1.6 mL O2/100 mL flow, while the cold group 3 showed 0.5 to 0.6 mL O2/100 mL flow (P < .001). Cold crystalloid cardioplegia (group 1) was similar to group 3 (0.3 mL O2/100 mL). All three groups were similar hemodynamically (cardiac output, cardiac index, left ventricular stroke work index). Two of 163 patients died in group 2. Four sustained stroke, three in group 2, one with a fatal outcome. CONCLUSIONS: Continuous warm cardioplegia (group 2) did not provide better myocardial protection despite that no CK-MB isoenzyme leak was demonstrated intraoperatively. Intermittent cold crystalloid cardioplegia and cold retrograde provided a clearer operative field and motionless heart. As long as O2 was adequately supplied, under 90 minutes' cross-clamp time, cold crystalloid cardioplegia and cold retrograde blood cardioplegia is safe under hypothermic conditions, whereas warm cardioplegia requires continuous uninterrupted technique with oxygen delivery.

Anatomical considerations of the venous drainage of the lower extremities: clinical implications.

The anatomical course, branches, valve distribution, and perforators of the long saphenous vein were described in 59 patients. Altogether, 95 legs were dissected while obtaining the vein for coronary artery bypass procedures. The findings indicated that the first valve is located 12.7 +/- 7.1 cm from the ankle, the second 21.1 +/- 7.4 cm, and the third 25.4 +/- 5.5 cm. The branches are located at 14.2 +/- 6.5 cm and the second is 20.4 +/- 7.9 cm. The perforators indirectly connected were present at 16.3 +/- 6.25 cm and 24.0 +/- 2.6 cm. Based on the anatomy of valves, branches, and perforators the areas of vulnerability and disease in the long saphenous system are localized in the area of 12 to 16 and 20 to 25 cm from the ankle. The most probable location of the first branch, perforator, or valve is between 12 to 16 cm from the ankle, probability of 0.70. The valve and the branch distributions of the long saphenous vein on the lower extremities are consistent. The "tier" arrangement of the venous drainage and the supporting role of the membranous fascia were also emphasized. These observations seem to be, in some respect, controversial to the general accepted anatomical descriptions.

The revival of the horseshoe graft (side-to-side saphenous-vein-to-aorta anastomosis).

The technique of the horseshoe graft, side-to-side saphenous-vein-to-aorta anastomosis, using a long saphenous vein harvested from the leg is described in this communication. The lack of valves between the ankle and the knee and the high caliber natural Y distributions make this technique attractive. Constructing a double set of vein graft provides, in particular, unlimited possibilities for the construction of six to eight grafts if this appears anatomically feasible.