Philip Ovadia

Platelet-activating factor (PAF) in experimental and clinical sepsis.

Despite considerable progress in understanding the pathogenic mechanisms of Gram-negative sepsis, the outcome of septic patients has not significantly improved. There are ample data that support a role for inflammatory mediators in sepsis that act in synergy with infectious agents to initiate and propagate the disease process. One such mediator is the glycerophospholipid platelet-activating factor (PAF). The objective of the present review is to summarize experimental and clinical evidence implicating PAF as a mediator in the pathomechanism of sepsis. This review is timely because many potent and selective PAF antagonists have matured for clinical development and a careful analysis of the data that support or refute the merit of clinical trials with such compounds may be important for both academic and pharmaceutical applications.

Chest Tube Removal: End-inspiration or End-expiration?

BACKGROUND Recurrent pneumothorax is the most significant complication after discontinuation of thoracostomy tubes. The primary objective of the present study was to determine which method of tube removal, at the end of inspiration or at the end of expiration, is associated with a lesser risk of developing a recurrent pneumothorax. A secondary objective was to identify potential risk factors for developing recurrence. METHODS A prospective study of 102 chest tubes in 69 trauma patients (1.5 tubes per patient) randomly assigned to removal at the end of inspiration (n = 52) or the end of expiration (n = 50). RESULTS Recurrent pneumothorax or enlargement of a small but stable pneumothorax was observed after the removal of four chest tubes in the end-inspiration group (8%) and after discontinuation of three chest tubes (6%) in the end-expiration group (p = 1.0). Of those, only two tubes in the end-inspiration group and 1 tube in the end-expiration group required repeat closed thoracostomy. Multiple factors were analyzed that did not adversely affect outcome. These included patient age, Injury Severity Score, Revised Trauma Score, mechanism of injury, hemothorax, thoracotomy, thoracostomy, previous lung disease, chest tube duration, the presence of more than one thoracostomy tube in the same hemithorax, or a small (but stable) pneumothorax at the time of tube removal. CONCLUSIONS Discontinuation of chest tubes at the end of inspiration or at the end of expiration has a similar rate of post-removal pneumothorax. Both methods are equally safe.

Locally Produced Tumor Necrosis Factor-α Mediates Interleukin-2Induced Lung Injury

Abstract Interleukin (IL)-2–induced microvascular lung injury is an experimental paradigm commonly used to investigate the pathogenesis of the adult respiratory distress syndrome. Since tumor necrosis factor-α (TNF-α) is known to induce such an injury in vivo and since TNF-α is involved in other models of lung injury, we postulated that it might also mediate pulmonary toxicity after IL-2 administration. The present study tested this hypothesis by evaluating the effect of TNF-α inhibition on IL-2–induced lung injury in the rat. Recombinant human IL-2 (106 U IV per rat, n=6) elevated lung water, myeloperoxidase activity, and protein accumulation in bronchoalveolar lavage fluid and induced tissue hypoxia. Also, IL-2 enhanced lung tissue TNF-α mRNA and peptide (1543±496 pg/g lung wet weight) localized to alveolar macrophages by in situ hybridization. In marked contrast, IL-2 failed to affect serum TNF-α, which remained at undetectable levels. Pretreatment with anti–TNF-α monoclonal antibody (25 mg/kg IV, n=7) or the TNF-α synthesis inhibitor rolipram (200 μg/kg IV, n=7) attenuated lung injury and reverted tissue hypoxia. Furthermore, TNF-α inhibition prevented the upregulation of lung tissue IL-1β, IL-6, cytokine-induced neutrophil chemoattractant, and E-selectin (ELAM-1) but not intercellular adhesion molecule-1 mRNAs in response to IL-2. These data imply that locally produced TNF-α mediates IL-2–induced lung inflammation and tissue injury and point to the potential utilization of TNF-α inhibitors in treating the pulmonary toxicity of IL-2 immunotherapy.

An improved clinically relevant sepsis model in the conscious rat

Objective To develop an improved small animal experimental paradigm that more closely mimics human sepsis. Design Prospective, randomized, controlled animal study. Setting Medical school research laboratory. Subjects Male Sprague-Dawley rats (280–320 g). Interventions We monitored the hemodynamic, hematologic, and biochemical consequences of abdominal sepsis produced by intraperitoneal implantation of a fibrin clot containing Escherichia coli in conscious, antibiotic-treated rats. Measurements and Main Results Similar to human sepsis, the implanted, infected clot (LD50 = 5–7 × 108 colony forming units/mL, n = 6) elevated cardiac index (>7% vs. sterile clot, p < .05, at 4 hrs), whereas mean arterial pressure and heart rate remained unaffected. The total peripheral resistance index and stroke volume index tended to decrease and increase, respectively. In contrast, an intravenous bolus injection of endotoxin (LD50 of E. coli lipopolysaccharide = 5.6 mg/kg, n = 7), the most commonly used sepsis model, induced profound hypodynamic responses manifested by a 27% decrease (vs. endotoxin vehicle, p < .01) in cardiac index, a 28% increase in the total peripheral resistance index (p < .01), and a 33% decrease in the stroke volume index (p < .01). The infectious peritonitis model also displayed dose-dependent thrombocytopenia (<61%, p < .05), leukopenia (<60%, p < .05), and mortality rate (50% at 5–7 × 108 colony forming units/mL, p < .05) with a minimally elevated serum tumor necrosis factor-&agr; level (145 vs. 12 ± 6 pg/mL in controls, p < .05). Conclusion This rodent model of antibiotic-treated, intra-abdominal infection features key characteristics of clinical sepsis. Although the hyperdynamic response observed in septic patients undergoing resuscitation was not clearly elicited, this paradigm better mimics clinical sepsis compared with the commonly used endotoxin model. Thus, utilization of this paradigm may provide additional opportunities to explore mechanisms of sepsis and to examine novel therapeutics.

Abdominal injuries associated with lumbar spine fractures in blunt trauma

BACKGROUND Specific analysis of the relationship between abdominal injuries and lumbar spine fractures has not yet been reported. METHODS A retrospective review of 258 blunt trauma patients with lumbar spine fractures treated between 1991 and 1996. RESULTS 26 patients sustained concomitant lumbar spine fractures and abdominal injuries. The mechanism of injury was motor vehicle collision (73%), pedestrian-struck (11%), fall (8%) and assault (8%) resulting in ISS, RTS and mortality of 27 +/- 4, 6.5 +/- 0.4 and 8%, respectively. Forty-four lumbar spine fractures were identified (1.7/pt) in association with splenic (54%), renal (41%), hepatic (32%) and small bowel (23%) injuries and no retroperitoneal involvement. Multilevel lumbar spine fractures were associated with a higher organ injury/fracture ratio compared with single level fractures (p < 0.01) including a twofold higher incidence of solid organ (spleen, liver and kidney) injury (p < 0.01). The level and type of fracture did not affect the incidence of total and individual organ injury. Patients with abdominal injuries were more severely injured mainly due to increased incidence of associated thoracic injuries although no significant difference in mortality was observed. CONCLUSION Abdominal injuries occurred only in the minority of blunt trauma patients with lumbar spine fractures. These injuries, which followed a similar distribution pattern as in blunt trauma in general, occurred most commonly due to motor vehicle collisions and in association with multilevel vertebral fractures. No correlation with fracture type or level was identified.

Candida Pericarditis: Clinical Profile and Treatment

BACKGROUND Candida pericarditis is a rare medical and surgical emergency which, unless treated, leads to impaired cardiac function and death. To facilitate early diagnosis, the clinical features of this condition should be identified. METHODS Twenty-five cases of Candida pericarditis reported in the last 30 years along with 1 new case were reviewed with regard to demographics, precipitating factors, diagnosis, treatment, and outcome. RESULTS The syndrome occurred in immunocompromised (73%), antibiotic-treated (62%), or postpericardiotomy (54%) patients. The clinical presentation was frequently subtle and nonspecific. Nevertheless, unexplained fever, an increasing cardiac shadow on chest roentgenogram, or the development of cardiac tamponade may be suggestive. Positive culture for Candida in pericardial fluid or histologic evidence of yeast forms in pericardial tissue establishes the diagnosis. A combination of pericardiocentesis followed by operative drainage and antifungal agents is the usual treatment. Untreated, Candida pericarditis is 100% lethal, whereas prompt diagnosis and treatment lead to cure (mean follow-up, 19 months). CONCLUSIONS Fever and evolving cardiac tamponade in immunocompromised or postpericardiotomy patients may be suggestive of Candida pericarditis; the presence of organisms in pericardial fluid is diagnostic. Pericardiocentesis followed by operative drainage and antifungal agents appears to be the treatment that is most likely to be curative.

Admission patterns of an urban level I trauma center

Because trauma admission and hospitalization patterns have profound effects on the organization and utilization of urban trauma-care systems, the objective of this study was to identify and analyze these patterns. As an example, admissions to an urban Level I trauma center were reviewed. Retrospective review of all 2029 trauma admissions to a Level I trauma center was conducted from 1993 to 1996. The result was that most trauma patients were young (40% <30 years of age) and male (74%). Mechanisms of injury were motor vehicle accident (36%), fall (27%), gunshot (17%), stab (7%), assault (6%), and swimming or diving accident (3%). Half of the patients were directly admitted from the scene. Injury Severity Score, length of stay, and mortality were 14.1 ± 0.3, 10.5 ± 0.3 days, and 5.1%, respectively. Admissions tended to occur more frequently between 4:00 PM and midnight (46%), between Friday and Sunday (52%), and between July and October (41%). The following patterns were identified: admissions per year decreased (-21%) because of reduced penetrating trauma (-43%, P < .01); pediatric patients (<15 years) had similar incidence of penetrating trauma as adults (ages 15-45). Length of stay for all mechanisms of injury was not statistically different; most mortalities occurred within the first day (33%, P < .01) or after 6 days (36%, P < .01); early mortality was mainly due to penetrating injury (74%, P < .01), whereas late mortality was related to blunt trauma (92%, P < .01). The conclusion was that admission and demographic patterns were identified, which may be useful in the utilization, modification, and future design of trauma systems.

Liposome-encapsulated hemoglobin does not exacerbate endotoxin-induced lung injury.

Objective To test the hypothesis that liposome encapsulated hemoglobin (LEH), an experimental oxygen-carrying fluid, exacerbates endotoxin-induced lung injury in the rat. Design Prospective, randomized animal study. Setting University animal laboratory. Methods Anesthetized Sprague-Dawley rats (n = 8–13) were infused with LEH (10% of estimated total blood volume) or vehicle (0.9% NaCl). Thirty minutes later, Escherichia coli endotoxin (3.6 mg/kg, iv) or vehicle (0.9% NaCl) was administered, and skeletal muscle oxygen tension as well as lung injury were assessed at 2, 4, and 8 hrs. Oxygen tension was measured using a miniaturized thin film oxygen sensor placed in the rectus abdominis muscle, and lung injury was evaluated by determining lung weights, lung myeloperoxidase activity, lung tissue tumor necrosis factor-&agr; level, and protein concentration in bronchoalveolar lavage fluid. Results The intravenous bolus injection of E. coli endotoxin elevated lung water content (33% ± 5%;p < .01 vs. sham controls), myeloperoxidase activity (56% ± 6%;p < .01), and tumor necrosis factor-&agr; production (1320 ± 154 pg/g lung tissue;p < .05 vs. undetected levels in sham controls), as well as induced protein accumulation in bronchoalveolar lavage fluid (258% ± 38%;p < .01) and skeletal muscle hypoxia (52 ± 8 mm Hg;p < .05). Pretreatment with LEH, which when infused alone did not induce lung injury, had no effect on these responses. Conclusion In this specific model of endotoxin-induced lung injury, LEH does not exacerbate microvascular leakage and leukosequestration, the hallmarks of adult respiratory distress syndrome.

Effect of Liposome-Encapsulated Hemoglobin on Triglyceride, Total Cholesterol, Low-Density Lipoprotein, and High-Density Lipoprotein Cholesterol Measurements

The present study investigated the effect of liposome-encapsulated hemoglobin (LEH), an experimental oxygen-carrying resuscitation fluid, on triglyceride, total cholesterol, and low density lipoprotein (LDL), and high density lipoprotein (HDL) cholesterol measurements. In vivo, the intravenous infusion of LEH (5.6 mL/kg, n=6) elevated serum triglycerides (+92% vs. baseline, P<.05), total cholesterol (+25% vs. baseline, P<.01), LDL cholesterol (+72% vs. baseline, P<.01) and had no effect on serum HDL cholesterol. In addition, LEH did not alter the elevation in serum triglycerides (+302% vs. baseline, P<.01) and LDL cholesterol (+86% vs. baseline, P<.01) induced by lipopolysaccharide (3.6 mg/kg, i.v., n=6). Ex vivo, measurements of triglycerides and total cholesterol as well as LDL and HDL cholesterol in whole blood from naive rats were not changed by the addition of LEH (0–50%, n=6). In vitro, the addition of a fixed concentration of LEH (50%, n=6) to varying concentrations of cholesterol solution (0–50%), or vice versa, had no effect on cholesterol determination. It is therefore concluded that LEH only minimally affects serum levels of triglycerides, total cholesterol, LDL cholesterol, and HDL cholesterol and does not interfere with their measurement.