Journal of Intensive Care Medicine

The role of percutaneous transluminal coronary angioplasty (PTCA) in the management of acute myocardial infarction (AMI) has not yet been precisely defined. The longest experience with PTCA in this setting has been in patients who are not candidates for thrombolytic therapy and in patients in whom thrombolysis has failed. Clinical interest has recently focused on direct use of PTCA (instead of thrombloysis) as the initial approach to reperfusion in AMI. We review the conceptual bases for both thrombolytic therapy and PTCA in AMI, and we then detail the clinical experience with PTCA in a variety of patient populations with AMI to guide use of both therapies in this setting.

Bone marrow necrosis is regarded as an uncommon entity that is associated with a poor prognosis. However, organized studies using either bone marrow biopsy specimens or autopsy material showed that bone marrow necrosis can be demonstrated in approximately one third of specimens. It is found in a large number of both malignant and nonmalignant disorders, in addition to occurring following large exposures to radiation or high dose cancer chemotherapy. In the absence of radiation or cancer chemotherapy, it probably eventuates from either vascular occlusion or blood stasis in small blood vessels. When bone marrow necrosis is prolonged, it may be associated with the development of bone marrow fibrosis, and it may serve as a predisposing etiology for idiopathic myelofibrosis. Most patients discovered with bone marrow necrosis have few symptoms, and they are eventually lost to follow-up without evident progression or development of a clinical illness. In acute disorders and in those who undergo effective therapy, recovery appears to occur without complications. This frequently overlooked finding is the subject of many case reports in the medical literature, but it has only been rarely systematically investigated. The latter is probably warranted because of the potential role of bone marrow necrosis in the pathophysiology of a number of disorders and the paucity of information for treatment of this pathological finding.

Aspiration pneumonia is a serious complication of enteral feeding. Many critically ill patients are particularly at risk for aspiration. Few studies have rigorously compared various access devices. Risk factors for aspiration and studies examining aspiration associated with enteral feeding devices are reviewed. We recommend a surgical jejunostomy for all patients at high risk for aspiration who require more than 3 weeks of enteral nutrition support.

We studied the biochemical and the clinical consequences of the application of continuous venovenous hemodiafiltration to the management of acute renal failure in critically ill patients. One hundred consecutive surgical and medical ICU patients with acute renal failure were entered into a prospective clinical study at an intensive care unit of tertiary institution. Intervention included assessment of illness severity by APACHE II score on admission and by organ failure score prior to initiation of renal replacement therapy; treatment of patients with continuous venovenous hemodiafiltration; and measurement of biochemical variables prior to and after therapy. Outcome assessment included incidence of complications, duration of oliguria, duration of intensive care and hospital stays, and survival to hospital discharge. Measurements and main results included the following: mean patient age was 60.9 years (range 21–81 yr); mean APACHE II score, 28.6 (95% confidence interval; 27.4-29.8); and number of failing organs, mean, 4.1 (95% confidence interval; 3.8-4.4). At commencement of continuous venovenous hemofiltration with dialysis, 79% of patients were receiving inotropic drugs and 72% were septic, and, in 35%, bacteremia or fungemia was demonstrated. Renal replacement therapy was applied for a mean duration of 186.2 hours (95% confidence interval; 149.2-223.7), with a mean hourly net ultrafiltrate production of 621 mL (95% confidence interval; 594–648) and a mean urea clearance of 28.1 mL/min (95% confidence interval; 26.7-29.5). Azotemia was controlled in all patients (plasma urea < 30 mmol/L). During the more than 18,000 hours of treatment, there was no therapy-associated hemodynamic instability. Complications were rare. They included two cases of hemofilter rupture with minor blood loss and a single case of bleeding at the site of the vascular-access catheter. Forty-three patients survived to ICU discharge, and 40 survived to hospital discharge. Continuous venovenous hemodiafiltration is a safe and an effective form of renal replacement therapy in critically ill patients. In such patients, who have a high predicted mortality rate, it was associated with a 40% survival rate. These findings suggests that continuous venovenous hemodiafiltration may be ideally suited to patients with multisystem organ failure with acute renal failure.

I describe the pathophysiological and hemodynamic events that occur after an emergency pneumonectomy for trauma and how they impact on subsequent mortality. Four patients were identified as requiring an emergency right pneumonectomy for trauma at a level 1 Urban Trauma Center within a 39-month period. A retrospective review of their hospital course served as the basis for our analysis. Three patients sustained gunshot wounds and one patient was a victim of blunt trauma. Hemodynamic data were available for three patients who survived more than 24 hours. All patients presented in shock and required massive transfusion. One patient died in the operating room due to air embolism and shock. After pneumonectomy, there was an increase in pulmonary artery pressure (PAP) and pulmonary vascular resistance (PVR) more than 2 times normal, which coincided with a decrease in stroke volume, cardiac output, and right and left ventricular stroke work (RVSW/LVSW). The RVSW gradually increased to above normal levels by postoperative day 5, whereas the LVSWI remained below normal. Adult respiratory distress syndrome (ARDS) developed in all patients early in the postoperative period. There was evidence of oxygen delivery (DO₂) dependent of oxygen consumption (VO₂) and the DO₂ remained below normal despite inotrope administration. The remaining three patients died 7 to 13 days after surgery due to various combinations of ARDS, cardiac failure, and sepsis. Until we have better methods to decrease PAP selectively, traumatic pneumonectomy should be avoided if possible, especially when it involves the right side or is associated with a contralateral lung injury. Early operative intervention and control of the pulmonary hilum may lessen the severity of shock. The hemodynamic changes that occur after pneumonectomy for trauma becomes additive in the presence of ARDS. This combination results in inadequate cardiac function, oxygen transport, and, ultimately, death.

The American College of Chest Physicians recently published a statement summarizing the results of the Consensus Conference on Mechanical Ventilation (Chest 1993; 104:1833-1859). In this statement, the physiology of mechanical ventilation was extensively reviewed, and 5 principles and 38 guidelines or recommendations on which consensus was reached were presented. The following critique (1) summarizes the findings of the conference; (2) comments on selected principles, recommendations, and physiological concepts presented in the statement; and (3) discusses the strengths and the limitations of the consensus conference format, as well as the errors and biases inherent in the policy-making process. The Consensus Conference statement is one of the clearest, most complete reviews of physiology and techniques of mechanical ventilation I have yet encountered. The suggested guidelines and recommendations clearly follow from the physiological principles discussed. Unfortunately, numerous problems with the consensus conference format and the almost total lack of controlled randomized studies limit the confidence with which these suggestions can be accepted. Our experts in respiratory care are remiss for not providing outcome-based or preference-based data, and the pulmonary and critical care community is remiss for not demanding that they do so before changing our approaches on the basis of their suggestions.