Hidehiko Kushi

Early hemoperfusion with an immobilized polymyxin B fiber column eliminates humoral mediators and improves pulmonary oxygenation

IntroductionThe objective of this study was to clarify the efficacy and mechanism of action of direct hemoperfusion with an immobilized polymyxin B fiber column (DHP-PMX) in patients with acute lung injury or acute respiratory distress syndrome caused by sepsis.MethodThirty-six patients with sepsis were included. In each patient a thermodilution catheter was inserted, and the oxygen delivery index and oxygen consumption index were measured. DHP-PMX was performed in patients with a normal oxygen delivery index and oxygen consumption index (> 500 ml/minute per m2 and >120 ml/minute per m2, respectively). The Acute Physiology and Chronic Health Evaluation II score was used as an index of the severity of sepsis, and survival was assessed after 1 month. The humoral mediators measured were the chemokine IL-8, plasminogen activator inhibitor-1, and neutrophil elastase (NE). These mediators were measured before DHP-PMX treatment, and at 24, 48, and 78 hours after the start of treatment. The arterial oxygen tension (PaO2)/fractional inspired oxygen (FiO2) ratio was measured before DHP-PMX treatment and at 24, 48, 72, 92, and 120 hours after the start of treatment.ResultsAll patients remained alive after 1 month. Before DHP-PMX treatment, the Acute Physiology and Chronic Health Evaluation II score was 24 ± 2.0, the IL-8 level was 54 ± 15.8 pg/ml, plasminogen activator inhibitor-1 was 133 ± 28.1 ng/ml, and NE was 418 ± 72.1 μg/l. These three humoral mediators began to decrease from 24 hours after DHP-PMX treatment, and the decline became significant from 48 hours onward. The PaO2/FiO2 ratio was 244 ± 26.3 before DHP-PMX treatment but improved significantly from 96 hours onward. There were significant negative correlations between the PaO2/FiO2 ratio and blood levels of NE and IL-8.ConclusionThe mechanism of action of DHP-PMX is still not fully understood, but we report the following findings. The mean blood levels of plasminogen activator inhibitor-1, NE, and IL-8 were significantly decreased from 48 hours after DHP-PMX treatment. The mean PaO2/FiO2 ratio was significantly improved from 96 hours after DHP-PMX treatment. Improvement in the PaO2/FiO2 ratio appeared to be related to the decreases in blood NE and IL-8 levels.

The Clinical Issue and Effectiveness of Brain Hypothermia Treatment for Severely Brain-Injured Patients

Treatment for brain injury has focused on neuroprotection against brain edema, brain ischemia, and control of intracranial hypertension in animal studies. However, these concepts are not successful in the management of in severely brain-injured patients, because animal studies do not include information about the influence of the excess response of systemic circulatory-metabolic changes and hypothalamus-pituitary endocrine axis dysfunction caused by anesthesia. In our recent clinical studies of severe brain injury, the management of restoration therapy of dying neurons in injured brain tissue before neuroprotection therapy produced successful clinical results. Three major targets of treatment are important: Adequate administration of oxygen and metabolic substrates, control of excess release of vasopressin and growth hormone for prevention of blood-brain barrier dysfunction and cytokine encephalitis, and preclusion of selective neuronal radical damage of the dopamine A 10 nervous system for prevention of vegetation. Management of intensive care is focused on three subsequent targets: Initial care management is to maintain sufficient cerebral oxygenation and adequate brain metabolism by control of PaO2/FiO2above 300, systolic blood pressure above l00 mmHg, serum glucose at 120–140mg/dl, brain temperature at 34°-32°C, ICP below 20 mmHg, antithrombin-III (AT-III) above 100%, hemoglobin 2,3 diphosphoglycerate at 12–15 mmol/mL, serum pH above 7.3, oxygen delivery above 800ml/min, and fluid resuscitation. The second target is control of hypothalamus-pituitary axis activation by management of brain tissue temperature at 34°C with 120–140 mg/d serum glucose, 9% saline followed by 4% saline fluid resuscitation, AT-III above 100% followed by replacement of albumin drip, and maintaining serum albumin higher than 3.5g/dl within 2–3 h after injury. The third target is to prevent the selective neuronal damage of the dopamine A 10 nervous system by controlling brain tissue temperature at 32°-33°C with management of Hb above 11 g/dl and administration of vitamin E and C. However, in brain hypothermia treatment, there are five major pitfalls: hemoglobin dysfunction associated with masking brain hypoxia, inadequate management of brain tissue temperature with hyperglycemia, undesirable duration of brain hypothermia, inadequate care management of systemic infections, and misunderstanding of nutritional issues in the management of brain injury. These pitfalls are discussed in this chapter. A novel technique for control of these clinical issues, the success of adequate neuronal oxygenation and brain metabolism, neurohormonal control of the blood-brain barrier dysfunction, and preservation of the dopamine A 10 nervous system are explained. Prevention of brain edema, elevation of ICP, and neuroexcitaion are not mechanisms of brain hypothermia. The effectiveness of brain hypothermia treatment was studied by comparing the clinical results of brain hypothermia (99 cases) and normothermia (65 cases) of head injury with Glascow Coma Scale (GCS) less than 6. Patients with initial GCS scores of 3 did not benefit from brain hypothermia treatment, but with GCS 4, 5, and 6, clinical benefit was observed. As clinical signs, memory, intelhgence, and personality were not much disturbed in the brain hypothermia-treated group.

L-8 is a key mediator of neuroinflammation in severe traumatic brain injuries

The subjects were 22 patients with severe head injury. The average age was 45 ± 18.3 years. There were 13 survivors and 9 fatalities. Samples of peripheral blood and cerebrospinal fluid (CSF) were taken four times, at the time of admission and at 24, 72, and 168 hours later. IL-6: For the survivor group, peripheral blood levels were 181, 105, 37, and 26 pg/ml, respectively (median values). CSF levels were 5376, 3565, 328, and 764 pg/ml, respectively. For the fatality group, peripheral blood levels were 102, 176, 873, and 3059 pg/ml, respectively, whereas CSF levels were 15241, 97384, 548225, and 366500 pg/ml, respectively. IL-8: For the survivor group, peripheral blood levels were 36, 15, 15, and 15 pg/ml, respectively, whereas CSF levels were 23736, 4074, 355, and 1509 pg/ml, respectively. For the fatality group, peripheral blood levels were 21, 28, 43, and 77 pg/ml, respectively, whereas CSF levels were 29003, 8906, 5852, and 8220 pg/ml, respectively. IL-6 and IL-8 levels were significantly higher after 72 hours in the fatality group. The fact that CSF IL-8 was 1000 times that in the peripheral blood at the time of admission, and decreased thereafter, indicates that IL-8 is a key mediator of neuroinflammation.

Hemoperfusion with an immobilized polymyxin B column reduces the blood level of neutrophil elastase.

Background: We investigated whether direct hemoperfusion with an immobilized polymyxin B column (DHP with PMX) could reduce the blood level of neutrophil elastase. Methods: 20 sepsis patients were enrolled in the study. DHP with PMX was performed twice within a 24-hour period. Neutrophil elastase was measured 7 times. Results: Neutrophil elastase was 468 ± 75.1 µg/l, while it was 1,531 ± 201.7 µg/l immediately after the first session, declined to 351 ± 73.9 µg/l before the second session of DHP with PMX, and increased again to 599.3 ± 112.7 µg/l immediately after the second session, 328 ± 73.7 µg/l at 24 h, 264 ± 39.3 µg/l at 48 h, and 230 ± 36.1 µg/l at 72 h after DHP with PMX. The levels from 48 h onwards were significantly lower compared with that before treatment. Conclusion: DHP with PMX has an overall effect that reduces circulating neutrophil elastase levels.

Hemoperfusion With an Immobilized Polymyxin B Fiber Column Inhibits Activation of Vascular Endothelial Cells

Abstract:  Involvement of the activation of neutrophils and vascular endothelial cells in the pathology of sepsis has recently been reported. We therefore investigated whether direct hemoperfusion (DHP) with a polymyxin B immobilized fiber column (PMX) could reduce the level of plasminogen activator inhibitor‐1 (PAI‐1), an index of vascular endothelial cell activation. Twelve sepsis patients satisfying the following criteria were enrolled in the study: (i) stable global oxygen metabolism (oxygen delivery index > 500 mL/min/m2 and oxygen consumption index > 120 mL/min/m2); (ii) abnormal tissue oxygen metabolism (PCO2 gap: gastric mucosal PCO2 minus arterial PCO2 difference > 8 mm Hg); and (iii) mean blood pressure ≥ 60 mm Hg. Direct hemoperfusion with PMX was performed twice (for 3 h each time) within 24 h. Plasminogen activator inhibitor‐1 was measured a total of 5 times: before PMX‐DHP, immediately after the first DHP with PMX session (3 h after the start), and 24, 48, and 72 h afterward. The PAI‐1 value was 150 ± 30.0 ng/mL before DHP with PMX, 178 ± 60.0 ng/mL immediately after DHP with PMX, 90 ± 22.1 ng/mL at 24 h after, 65 ± 21.0 ng/mL at 48 h after, and 64 ± 18.3 ng/mL 72 h after. The values were significantly lower from 48 h onward compared with baseline. These data suggest that DHP with PMX inhibits vascular endothelial cell activation.

Gadolinium DTPA-Enhanced Magnetic Resonance Imaging of Cerebral Contusions

The morphological characteristics of cerebral contusions in head trauma patients suggest that an increase in cerebrovascular permeability is responsible for the contusion edema which develops within 1-3 days posttrauma. In the present study, 10 patients with cerebral contusions (mean age, 38 years old; 8 males and 2 females) were examined by gadolinium (Gd)-DTPA enhanced magnetic resonance imaging (MRI) at 1-2 days after trauma. Gd-DTPA (0.3 mmol/kg) was infused intravenously over a period of 30 min. MRIs were taken before, and at 2 and 4 hours after initiation of the Gd-DTPA administration. It was found that contusion edema areas were frequently enhanced by Gd-DTPA at 2 hours. The enhancement diminished at 4 hours. These findings appear to be inconsistent with the results of previously reported similar studies in which enhancement was detected at 6-9 days posttrauma but not during the period earlier than 6 days. This discrepancy may be attributable to the presence of a high blood concentration of Gd-DTPA for a longer period of time and a delay in the time at which MRIs were taken in the present study. The present data indicate that an increased cerebrovascular permeability occurs at as early as 1-2 days posttrauma, and suggest that contusion edema which progresses during the initial 1-3 days may be at least partially vasogenic in nature.

Hemoperfusion With an Immobilized Polymyxin B Fiber Column Improves Tissue Oxygen Metabolism

Abstract:  Recently, direct hemoperfusion with a polymyxin B‐coated fiber column (DHP‐PMX) has been increasingly used for the treatment of sepsis, and an improvement in outcomes has been reported. However, the mechanism of the method is not known in detail. In the present study, we examined whether the performance of DHP‐PMX improved tissue oxygen metabolism in patients with sepsis. Twenty‐two patients with sepsis, satisfying the following criteria, were enrolled in the study: (i) signs of systemic inflammatory response syndrome caused by infection; and (ii) mean arterial blood pressure ≥60 mm Hg (irrespective of the use of catecholamines). A thermodilution catheter was inserted prior to DHP‐PMX for appropriate intravenous infusion, and the DHP‐PMX was carried out twice within 24 h (for 3 h each time). Then, the gastric mucosal‐arterial PCO2 difference (PCO2 gap) was calculated as the gastric mucosal PCO2 minus arterial PCO2. A PCO2 gap ≥8 mm Hg was used to define abnormal tissue oxygen metabolism. PCO2 gap was measured before PMX‐DHP, as well as 24, 48, and 72 h afterward. PCO2 gap was 20 ± 4.9 mm Hg before DHP‐PMX vs. 16 ± 2.7 mm Hg (P = 0.189) at 24 h, 12 ± 2.8 mm Hg (P = 0.046) at 48 h, and 11 ± 2.2 mm Hg (P = 0.045) at 72 h afterward, showing a significant decrease from 48 h onward compared with before treatment. These findings suggest that DHP‐PMX improves tissue oxygen metabolism.

The risk factors for the occurrence of acute brain swelling in acute subdural hematoma

There are many cases of acute subdural hematoma (ASDH) in which acute brain swelling (ABS) occurs intra- or post-operative period and the control of intracranial pressure becomes impossible, resulting in death. Consequently, we investigated the risk factors for ABS in ASDH. Thirteen cases were used for this study. The average age was 53.8 years, and all the subjects were males whose Glasgow Coma Scale (GCS) was an average of 4.8 when hospitalized. We examined the relationship of blood gas values (pH, PaO2, PaCO2, HCO3-, and base excess) when hospitalized, mean arterial blood pressure (MAP) before surgery, and the presence of hypotension during the surgery with ABS. There were no cases, which showed evidence of hypoxia or hypotension when hospitalized or before the surgery. No statistical differences in blood gas values when hospitalized were observed between the 7 cases with ABS and the 6 cases without ABS. However, all of 7 cases with ABS showed MAP > or = 140 mmHg, and 6 cases demonstrated intraoperative hypotension (systolic pressure < 90 mmHg). Conspicuous hypertension before the surgery was shown to be a possible risk factor for ABS. Therefore; MAP > or = 140 mmHg may be a warning sign for the occurrence of ABS.

Importance of Metabolic Monitoring Systems as an Early Prognostic Indicator in Severe Head Injured Patients

We have analyzed and evaluated what is the best metabolic monitoring system to determine the prognosis for maintenance of neurological function in severe head injured patients. Acute subdural hematoma (ASDH) was recognized in fifteen of 22 patients and cerebral contusion in seven in this series. Intracranial pressure (ICP), jugular venous pH and jugular bulb venous oxygen saturation (SjO2) were continuously monitored as soon as possible following stabilization. The measurement of cerebral blood flow (CBF) was carried out using a stable Xenon-computerized tomography (Xe-CT). After measuring CBF, 3% carbon dioxide (CO2) loading was conducted to determine CO2 responses (delta CBF/delta CO2). In patients who died (D), jugular venous pH showed evidence of acidosis (6.3-7.2) with delta CBF/delta CO2 < 1 and cerebral metabolic rate of oxygen (CMRO2) < 1.21 within several hours of the trauma. On the other hand, arterial pH was shown to be within the normal range. In vegetative state (VS) and severe disability (SD) patients, jugular venous pH was shown to be within normal range, with delta CBF/delta CO2 < 1 and 1.44 < CMRO2 < 1.79. In all of moderate disability (MD) and good recovery (GR) patients, jugular venous pH was shown to be within the normal range, with delta CBF/delta CO2 > 1 and 1.65 < CMRO2 < 1.85. These results suggest that jugular venous pH, CO2 response and CMRO2, were useful as early prognostic indicators in the maintenance of neurological function.

Hemoperfusion with an immobilized polymyxin B fiber column reduces circulating interleukin-8 concentrations.

Abstract:  In the present study, we examined whether the performance of hemoperfusion with an immobilized polymyxin B fiber column (DHP‐PMX) reduces circulating interleukin‐8 concentration in patients with sepsis. Fifteen patients with sepsis satisfying the following criteria were enrolled in the study: (i) signs of systemic inflammatory response syndrome caused by infection; and (ii) mean arterial blood pressure ≥60 mm Hg (irrespective of the use of catecholamines). A thermodilution catheter was inserted prior to DHP‐PMX for appropriate intravenous infusion, and the DHP‐PMX was carried out twice at 24 h intervals (for 3 h each time). Circulating interleukin‐8 concentration was measured seven times. The sequential organ failure assessment (SOFA) score was calculated twice. Circulating interleukin‐8 concentration was 55 ± 15.7 pg/mL before DHP‐PMX, while it was 101 ± 58.8 pg/mL immediately after the first session of treatment. It was 24 ± 9.0 pg/mL before the second session of DHP‐PMX, and it was 28 ± 8.0 pg/mL immediately after the second session. The IL‐8 level was 17 ± 4.3 pg/mL at 48 h afterward, and 18 ± 4.3 pg/mL at 72 h afterward, showing a significant decrease from 48 h onwards, compared with before treatment (P < 0.05). The SOFA score was 9 ± 1.5 and the APACHE II score was 19 ± 2.0 before DHP‐PMX, while the SOFA score was 7.0 ± 0.9 at 72 h afterward, showing a significant decrease compared with before treatment (P < 0.05). The present findings indicate that DHP‐PMX indirectly reduces circulating interleukin‐8 concentration and improves SOFA score.