Partial left heart bypass for thoracic aorta repair. Survival without paraplegia.

OBJECTIVE: Emergency repair of the torn descending thoracic aorta has been associated with an almost 15% incidence of paraplegia. The literature to date suggests that the incidence of paraplegia is not influenced by mechanical adjuncts to enhance distal aortic perfusion during cross-clamping and therefore, "clamp and sew" has been considered an acceptable technique. The purpose of the present study was to review our experience with repair of descending thoracic aortas using partial left heart bypass and to compare this favorable initial experience with the available data on the use of the heparinless centrifugal pump. DESIGN: A retrospective review of the routine use of partial left heart bypass in 16 consecutive patients with descending thoracic aorta disruptions. Results were compared with similar reports in the recent literature on trauma. SETTING: A level 1 trauma facility in the Denver, Colo, metropolitan area. PATIENTS: All patients with multisystem blunt trauma with a mean injury severity score of 36. INTERVENTION: Repair of the descending thoracic aorta disruption using partial left heart bypass with a heparinless centrifugal pump. MAIN OUTCOME MEASURES: Primary outcome measures were survival and paraplegia; other monitored variables included proximal and distal aortic pressure, flow rates, and oxygen transport. RESULTS: Among the 14 survivors (88%) there were no cases of paraplegia, and intraoperative hemodynamics and oxygen transport were well maintained with partial left heart bypass. This experience is added to the available reported data on the use of the centrifugal pump. In these additional 42 patients, the mortality rate was 7%, with no incidence of paraplegia. CONCLUSIONS: This collective experience failed to disclose a single case of paraplegia when partial left heart bypass was employed for repair of descending thoracic aorta. Moreover, the use of partial left heart bypass in this cohort of critically injured patients is associated with survival and perioperative morbidity rates comparable with the best recent reports of emergency thoracic aortic repair.

Early predictors of postinjury multiple organ failure.

OBJECTIVE: To find a predictive model for postinjury multiple organ failure (MOF). DESIGN: A 3-year cohort study ending December 1992 (first year: retrospective; last 2 years: prospective). SETTING: Denver General Hospital (Colo) is a regional level I trauma center. PATIENTS: Consecutive trauma patients with an Injury Severity Score (ISS) greater than 15, with an age greater than 16 years, and who survived longer than 24 hours. Stepwise logistic regression analysis was performed in all patients (n = 394), in the subgroup of patients with 0 to 12 hours, plus 12 to 24 hours base deficit (BD) results (n = 220), and in a second subgroup of patients with BD plus lactate results at 0 to 12 hours and 12 to 24 hours (n = 106). MAIN OUTCOME: Postinjury MOF. RESULTS: The following variables were identified as independent predictors of MOF in the analysis of all patients: age more than 55 years, ISS greater than or equal to 25, and more than 6 U of red blood cells in the first 12 hours after admission (U RBC/12 hours). In the subgroup with BD results, the same analysis identified age greater than 55 years, greater than 6 U RBC/12 hours, and BD greater than 8 mEq/L (0 to 12 hours), while in the last subgroup analysis including BD and lactate results, greater than 6 U RBC/12 hours, BD greater than 8 mEq/L (0 to 12 hours), and lactate greater than 2.5 mmol/L (12 to 24 hours) were independently associated with MOF. CONCLUSIONS: Age greater than 55 years, ISS greater than or equal to 25, and greater than 6 U RBC/12 hours are early independent predictors of MOF. Subgroup analyses indicate that BD and lactate levels may add substantial predictive value. Moreover, these results emphasize the predominant role of the initial insult in the pathogenesis of postinjury MOF.

The first randomized trial of human polymerized hemoglobin as a blood substitute in acute trauma and emergent surgery.

BACKGROUND: Human polymerized hemoglobin (PolyHeme) is a universally compatible, disease-free, oxygen-carrying resuscitative fluid. This is the first prospective, randomized trial to compare directly the therapeutic benefit of PolyHeme with that of allogeneic red blood cells (RBCs) in the treatment of acute blood loss. STUDY DESIGN: Forty-four trauma patients (33 male, 11 female) aged 19-75 years with an average Injury Severity Score (ISS) score of 21+/-10 were randomized to receive red cells (n = 23) or up to 6 U (300 g) of PolyHeme (n = 21) as their initial blood replacement after trauma and during emergent operations. RESULTS: There were no serious or unexpected adverse events related to PolyHeme. The PolyHeme infusion of 4.4+/-2.0 units (mean +/- SD) resulted in a plasma [Hb] of 3.9+/-1.3 g/dL, which accounted for 40% of the total circulating [Hb]. There was no difference in total [Hb] between the groups before infusion (10.4+/-2.3 g/dL control vs. 9.4+/-1.9 g/dL experimental). At end-infusion the experimental RBC [Hb] fell to 5.8+/-2.8 g/dL vs. 10.6+/-1.8 g/dL (p < 0.05) in the control, although the total [Hb] was not different between the groups or from pre-infusion. The total number of allogeneic red cell transfusions for the control and experimental groups was 10.4+/-4.2 units vs. 6.8+/-3.9 units (p < 0.05) through day 1, and 11.3+/-4.1 units vs. 7.8 +/-4.2 units (p = 0.06) through day 3. CONCLUSIONS: PolyHeme is safe in acute blood loss, maintains total [Hb] in lieu of red cells despite the marked fall in RBC [Hb], and reduces the use of allogeneic blood. PolyHeme appears to be a clinically useful blood substitute.

Hypoxic events in the surgical intensive care unit.

An oxygen-monitoring protocol was established in the surgical intensive care unit (SICU) at the Denver General Hospital in July 1988. A 3-month surveillance audit ending March 1989 prospectively documented 100 consecutive hypoxic events in 51 of 241 (21%) SICU patients. These episodes occurred during mechanical ventilation in 46 patients, during spontaneous ventilation in 15 patients with artificial airways, and the remaining 39 occurred in nonintubated patients. Hypoxemia was recognized by pulse oximetry in 59, arterial blood gas analysis in 24, mixed venous oximetry in 15, and transcutaneous oxygen monitoring in 2. These events were due to problems with the ventilator or airway in 42, recent interventions in 21, new pulmonary process in 19, progression of underlying disease in 11, and unknown causes in 7. Two thirds resulted from mechanical problems amenable to simple intervention; there were two adverse outcomes. In conclusion, acute hypoxia is a frequent potentially morbid SICU event. Advances in continuous oxygen monitoring permit early identification and thereby may limit adverse outcomes, but should not prompt an expensive diagnostic work-up.

Pneumonia: cause or symptom of postinjury multiple organ failure?

Recent studies have shown that selective gut decontamination can reduce the incidence of pneumonia, but this does not decrease multiple organ failure (MOF) or mortality. These findings have prompted the hypothesis that pneumonia is an inconsequential symptom of MOF. To test this, we prospectively evaluated 123 high-risk trauma patients (mean Injury Severity Score = 36.2 +/- 1.5). Organ dysfunction, scored daily according to a 12-point scale, ultimately developed in 28 (23%) patients. Major infections were diagnosed, based on strict criteria, in 59 patients (48%), and pneumonia developed in 52 patients (43%). Pneumonia was significantly associated with MOF (82% of patients with MOF versus 30% of patients without MOF, p < 0.0001). In 14 (50%) of the patients with MOF, pneumonia preceded a significant rise (greater than or equal to 3) in serial MOF scoring. Of note, 10 (71%) of these patients died. Among the remaining 14 patients with MOF, 10 developed pneumonia, but this was associated with a minimal increase (less than or equal to 2) in MOF scoring (3 patients died). These data, by temporal association with MOF scoring, implicate pneumonia in precipitating or significantly worsening organ failure in 50% of the patients who developed MOF.

Efficacy of selective intrabronchial air insufflation in acute lobar collapse.

Flexible fiberoptic bronchoscopy (FFB) to remove mucous plugs followed by selective intrabronchial air insufflation (SII) to expand the atelectatic lung was used in 17 surgical intensive care unit (SICU) patients with pulmonary lobar collapse. Thirteen patients were admitted for acute trauma, and the remainder were elderly postoperative patients. Lobar collapses occurred on SICU days 1 to 18 (mean +/- SEM: 5 +/- 1 days), and duration ranged from 4 to 258 hours (mean: 77 +/- 18 hours). Indications for FFB with SII included critical hypoxemia in 5 patients, worsening collapse in 2, and failure to respond to aggressive respiratory care in 10 (59%). FFB with SII was effective in 14 (82%) patients: 10 achieved full lung re-expansion, and 4 partial lung re-expansion. When lobar collapse was of less than 72 hours' duration, 92% (12 of 13) of patients had lungs re-expanded compared with 50% (2 of 4) whose collapse existed for more than 72 hours. The mean PaO2/FIO2 (fraction inspired oxygen) ratio was 135 +/- 18 prior to FFB with SII and increased to 205 +/- 21 after FFB with SII. Complications were minor and clinically insignificant. In conclusion, SII appears to be a simple, safe, effective adjunct to FFB in the treatment of SICU patients with pulmonary lobar collapse.

Optimal management of complicated empyema.

BACKGROUND: Despite continued improvement in medical therapy, empyema remains a challenging problem for the surgeon. Multiple treatment options are available; however, the optimal therapeutic management has not been elucidated. METHODS: A retrospective review was performed of all adult patients admitted to Denver Health Medical Center between January 1, 1993, and December 31, 1998, with the diagnosis of empyema. Data tabulated included patient demographics, presentation, chest computed tomography (CT) findings, treatment, and outcome. RESULTS: Empyema was diagnosed in 58 patients, 45 cases of which were multiloculated at the time of presentation. Empyema was secondary to pneumonia is 41 patients and posttraumatic in 15. In addition to antibiotic therapy, initial treatment included chest tube drainage alone (n = 6), chest tube drainage with primary operation (n = 19), and chest tube drainage with intrapleural fibrinolytic therapy (n = 33). In 15 patients (45%), fibrinolytic therapy failed. Initial chest CT revealed a pleural peel in 5 patients treated with fibrinolytics and all failed. Multiloculation, however, was not a factor in failure of fibrinolysis. Moreover, chest CT missed the presence of a pleural peel in 17 of 31 patients documented to have a significant peel at the time of thoracotomy. CONCLUSION: Multiple therapeutic options are available for the management of empyema. Multiloculation is not a contraindication to an initial trial of chest tube drainage or fibrinolytic therapy. In contrast, CT evidence of a pleural peel uniformly predicted failure of nonoperative treatment.

Jack A. Barney Resident Research Award. Cardiopulmonary hazards of perihepatic packing for major liver injuries.

BACKGROUND: Perihepatic packing has been shown to result in pathologic intra-abdominal hypertension. Although now recognized as impairing abdominal organ perfusion, the extent to which perihepatic packing affects cardiopulmonary function has not been elucidated. METHODS: We analyzed a 3-year experience with 11 patients who sustained major hepatic injuries requiring perihepatic packing to control hemorrhage. Pertinent hemodynamic indices consisting of pulmonary capillary wedge pressure (PCWP), cardiac index (CI), oxygen delivery index (DO2), and systemic vascular resistance (SVR), and pulmonary indices consisting of peak airway pressure (PAP), mean airway pressure (MAP), static compliance (CST), and PaO2/FiO2 were measured in the surgical intensive care unit immediately before and after packs were removed. RESULTS: Unpacking resulted in a significant increase in CI (3.1 +/- 0.4 to 4.2 +/- 0.6 L/min/m2), DO2 (539 +/- 41 to 689 +/- 43 mL min/m2), CST (26 +/- 6 to 36 +/- 4 mL/cm H2O), and PaO2/FiO2 (162 +/- 44 to 237 +/- 53 cm H2O), as well as a significant decrease in PAP (47 +/- 9 to 29 +/- 6 cm H2O), MAP (34 +/- 4 to 27 +/- 3 cm H2O), PCWP (21 +/- 4 to 13 +/- 3 mm Hg), and SVR (1,239 +/- 162 to 887 +/- 130 dyne/cm5). CONCLUSIONS: Abdominal compartment syndrome following temporary perihepatic packing can result in significant cardiopulmonary compromise. While perihepatic packing can be an early life-saving procedure, timely alleviation of the secondary syndrome may be critical to the ultimate salvage of patients with marginal cardiopulmonary reserve.

Are corticosteroids salvage therapy for refractory acute respiratory distress syndrome?

BACKGROUND: Late acute respiratory distress syndrome (ARDS), characterized by progressive pulmonary interstitial fibroproliferation, is associated with mortality > 80%. Although previous large prospective trials failed to show a benefit of steroids in early ARDS, recent small reports describe improved survival in patients with late ARDS. Recognizing the pathogenetic differences between early and late ARDS, we employed steroid therapy in patients with refractory late ARDS. PATIENTS AND METHODS: Over a 5-year period, we treated 6 patients who were dying of isolated refractory ARDS with methylprednisolone sodium succinate (1 to 2 mg/kg every 6 hours). Ventilatory parameters and lung injury scores were serially recorded, and steroids were weaned based on clinical response. RESULTS: Steroids were instituted after 16 days of advanced mechanical ventilatory support. By day 7 of steroid therapy, there was clinically significant improvement in PaO2/FiO2 ratios (84 to 172) and lung injury scores (3.6 to 2.9); 5 patients (83%) survived. CONCLUSIONS: Steroid therapy appears to be effective in patients with refractory late ARDS. Prospective trials are needed to define the indications, timing of intervention, dose and duration, and precautions of steroid therapy.

Resuscitation of the injured patient with polymerized stroma-free hemoglobin does not produce systemic or pulmonary hypertension.

BACKGROUND: Hemoglobin-based blood substitutes appear poised to deliver the promise of a universally compatible, disease-free alternative to banked blood. However, vasoconstriction following administration of tetrameric hemoglobins has been problematic, likely because of nitric oxide binding. Polymerized hemoglobin is effectively excluded from the abluminal space because of its size, and is thus less likely to perturb vasorelaxation. We therefore hypothesized that hemodynamic responses would be no different in injured patients receiving polymerized hemoglobin versus banked blood. METHODS: Injured patients requiring urgent transfusion were randomized to receive either polymerized hemoglobin or banked blood. Systemic arterial pressure, pulmonary arterial pressure, cardiac index, pulmonary capillary wedge pressure, systemic vascular resistance, and pulmonary vascular resistance were measured serially. RESULTS: There was no difference in any of the measured hemodynamic parameters between patients resuscitated with polymerized hemoglobin versus blood. CONCLUSIONS: Polymerized hemoglobin given in large doses to injured patients lacks the vasoconstrictive effects reported in the use of other hemoglobin-based blood substitutes. This supports the continued investigation of polymerized hemoglobin in injured patients requiring urgent transfusion.

Tracheal gas insufflation is a useful adjunct in permissive hypercapnic management of acute respiratory distress syndrome.

BACKGROUND: Despite numerous advances in critical care, the mortality of postinjury acute respiratory distress syndrome (ARDS) remains high. Recently, permissive hypercapnia (PHC) has been shown to be a viable alternative to traditional ventilator management in patients with ARDS. However, lowering tidal volume, as employed in PHC, below 5 cc/kg impinges upon anatomic dead space and precipitates a significant rise in PaCO2 The purpose of this study was to determine if continuous tracheal gas insufflation (cTGI) is a useful adjunct to PHC by lowering PaCO2, thus allowing adequate reduction in minute ventilation to achieve alveolar protection. METHODS: Over a 5-year period, 68 trauma patients with ARDS were placed on permissive hypercapnia. Nine of these patients additionally received cTGI at 7 L/min. Arterial blood gas determinations and ventilatory parameters were examined immediately prior to the implementation of cTGI and after 6h. RESULTS: The cTGI produced significant improvement in pH (7.25 +/- 0.03 to 7.33 +/- 0.03), PaCO2 (72 +/- 5 to 59 +/- 5 torr), tidal volume (7.9 +/- 0.6 to 7.2 +/- 0.6 cc/kg), and minute ventilation (13 +/- 1 to 11 +/- 1 L/min; P < 0.05). CONCLUSIONS: Continuous TGI is a useful adjunct to permissive hypercapnia, allowing maintenance of an acceptable pH and PaCO2 while allowing further reduction in tidal volume and minute ventilation.

Alternatives to Swan-Ganz cardiac output monitoring.

When we are concerned about the adequacy of peripheral oxygen delivery, our monitoring efforts focus on cardiac output as the component easiest to manipulate. Thermodilution is the current method of determining cardiac output because of logistic convenience, but by no means should it be considered the gold standard. Thoracic electrical bioimpedance is an appealing alternative because of its simplicity, noninvasiveness, and ability to track physiologic trends, but there have been a number of warnings against relying on it exclusively, at least until it is further developed. Doppler ultrasound appears to be a valid method in skilled hands, but its accuracy in quantitating cardiac output, especially in the critically ill, remains questionable. Mixed venous oximetry and transcutaneous oxygen monitoring are promising but are still being evaluated.

Pulmonary consequences of severe chest trauma.

Combined flail chest and pulmonary contusion is a frequent problem in patients with blunt multisystem trauma admitted to the intensive care unit. These patients are at high risk for pneumonia and adult respiratory distress syndrome, which adds substantially to their morbidity and mortality rates. This article discusses the epidemiology and pathophysiology of this condition and the role of the respiratory care practitioner in the optimal management of these critically injured patients.

Advances in oxygen monitoring of trauma patients.

Monitoring the critically injured patient is imperative, to assure adequate resuscitation from shock. Unfortunately, the commonly monitored variables correlate poorly with ultimate survival. As a result, therapy is inappropriately focused. Invasive monitoring (by way of a pulmonary artery catheter and arterial cannula) permit serial determinations of parameters pertaining to oxygen delivery as well as oxygen consumption. These are crucial in defining the shock state as well as the need for therapeutic intervention. Recent advances in mixed venous oximetry offer alternative means of assessing the adequacy of peripheral delivery of oxygen. With the advent of metabolic carts, it became feasible to measure respiratory gas exchange to determine oxygen consumption. Complexity, expense, and time clearly limit practical application of that technology to a small percentage of patients in the intensive care unit. Unfortunately, unrecognized hypoxemia remains a common problem. Advances in noninvasive monitoring offer alternative means to assess oxygenation. Pulse oximetry and transcutaneous oxygen monitoring are the state of the art. The pertinent devices are easy to use, portable, and accurate. Knowledge of their technical and physiologic limitations is needed to assure reliability. Their potential role extends beyond the intensive care unit setting, such as with prolonged radiologic evaluation or difficult transportation.

Clinical benefits of early post-injury enteral feeding.

Over the past two decades, clinical studies have provided convincing evidence that early nutritional support benefits metabolically stressed surgical patients by preventing acute protein malnutrition. However, the optimal route of substrate delivery (ie, enteral versus parenteral) continues to be debated. Recent basic and clinical investigation offers the exciting possibility that the beneficial effects of traditional nutritional support can be amplified by supplementing specific nutrients that exert pharmacological immune-enhancing effects. Over the past 15 years, the Department of Surgery at the Denver General Hospital has focused clinical research efforts on defining optimal nutrition following major torso trauma. The purpose of this paper is to review our studies as well as other clinical studies in order to answer three questions: 1) Does early post-injury nutritional support improve patient outcome? 2) What is the preferred route of substrate delivery? 3) Do new 'immune-enhancing' diets offer additional clinical benefits?

Auto-PEEP in the multisystem injured patient: an elusive complication.

Auto-PEEP (A-PEEP), unrecognized alveolar positive and expiratory pressure during mechanical ventilation, is an acknowledged hazard in patients with chronic obstructive lung disease. We evaluated 50 consecutive trauma patients for the presence of A-PEEP and its effect on hemodynamic stability. Injury Severity Scores (ISS) were 8 to 41 (21 +/- 1); Revised Trauma Scores (RTS) ranged from 2.0 to 7.8 (6.2 +/- 0.2). Mode of ventilation was assist control, inspiratory flow rates were 40 to 120 L/M (78 +/- 2). A-PEEP, determined in the non-assisting patient by occluding the expiratory port at end exhalation, was present in 28 patients (56%) and ranged from 1 to 12 cm H2O (5.3 +/- 0.4 cm H2O). Data segregated by A-PEEP versus no A-PEEP were as follows (Mean +/- SEM): [table: see text] *P less than 0.05, VE = minute ventilation, Paw = mean airway pressure. Upon reversal of A-PEEP in the eight patients with levels greater than 5 cm H2O, mean blood pressure rose from 90 +/- 17 to 102 +/- 22 mm Hg and central venous pressure fell from 13 +/- 5 to 7 +/- 5 mm Hg. A-PEEP was successfully treated in these eight patients by increasing peak flows, minimizing VE requirements and selective use of bronchodilators. In sum, the hypermetabolic ventilated trauma patient should be monitored routinely for this common phenomenon which may have profound cardiopulmonary effects in the setting of acute resuscitation.

Percutaneous tracheostomy/gastrostomy in brain-injured patients--a minimally invasive alternative.

Tracheostomy and gastrostomy are frequent adjunctive procedures required in the management of patients with severe brain injuries to facilitate neurorehabilitation. We therefore evaluated the use of two minimally invasive surgical procedures, percutaneous tracheostomy (PT) and percutaneous endoscopic gastrostomy (PEG), in 27 patients with severe brain injuries. The mean age was 41 +/- 4 years, and 23 (85%) were men. All patients were intubated, and 19 (70%) required mechanical ventilator support on the day of PT/PEG. The endotracheal tubes had been in place for 1 to 21 days (mean, 8.7 +/- 0.8). All patients were stable from their acute brain injury; 13 had intracranial pressure (ICP) monitors in place. The Seldinger technique, as described by Ciaglia, was employed for PT. Following PT, a PEG was inserted by a modification of the Sachs-Vine "push" technique. We were uniformly successful in placing these access tubes. Complications were minor and not clinically significant. Three of 13 patients (23%) with ICP monitors had a transient rise in ICP related to PT and one of these patients developed local subcutaneous emphysema. Another patient experienced a mild cellulitis at the tracheostomy site. Of note, there were no PEG-related complications. In conclusion, PT and PEG are readily learned, minimally invasive procedures. In our experience with patients with severe brain injuries combined PT/PEG is a uniformly safe alternative to gain long-term access to the airway and gut.

Incommensurate oxygen consumption in response to maximal oxygen availability predicts postinjury multiple organ failure.

Untreated flow-dependent oxygen consumption (VO2) has recently been implicated as an unrecognized risk factor for multiple organ failure (MOF). We therefore prospectively studied 39 severely injured patients with known risk factors for multiple organ failure who were subjected to an established resuscitation protocol aimed at maximizing oxygen delivery (DO2 greater than 600 mL/min.m2) to attain a VO2 goal of greater than 150 mL/min.m2. Fifteen (38%) of these high risk patients did not meet this VO2 goal by 12 hours. These nonresponding patients had significantly elevated lactate levels, suggesting defective aerobic metabolism. Of note, this blunted VO2 response despite maximal efforts to enhance peripheral oxygen availability predicted MOF. These data serve to re-emphasize the importance of the initial shock insult in causing or priming the host for the development of late MOF.