In Brief

Article Outline

• Copyright

Surgeons are becoming increasingly involved in critical care. Surgical intensive care units (ICUs) are often directed or staffed by surgeons with critical care credentials. A growing number of young surgeons seek added qualifications by spending 1 to 2 years in a critical care fellowship after surgical residency. Aspects of care to the severely ill patient, which were traditionally handled by consultants, such as mechanical ventilation, renal failure, nutrition, or hemodynamic support, are now managed in the ICU by either the same surgeon who performed the operation or by a different surgeon who runs the ICU. The field of surgical critical care, full of uncertainties and in need of better evidence, lends itself to change and innovation. Advances in surgical critical care occur fast and often are short-lived or even contradictory. Surgeons understand deeply the anatomic and physiologic features of the human body and will hopefully play a major role in organizing, streamlining, and improving the way we deliver critical care to surgical patients.

The need to transfer patients out of the safe ICU environment is fraught with problems. Complications during transport are common and often disastrous. Special protocols have been designed and transfer teams assembled to make the travel safer. To avoid the risk of transfer to the operating room and back, some procedures can be offered at the bedside. Percutaneous tracheostomy, percutaneous endoscopic gastrostomy, and percutanous inferior vena cava filter placement are nowadays routinely performed at the ICU bedside. These practices have been shown to be cost-effective, as they save operating room time and personnel. It allows these procedures to be performed timely, safely, and rapidly. In 1 study, the difference in cost between performing percutaneous tracheostomies, gastrostomies, and vena caval filter placements at the bedside and the operating room was $611,944 over a period of 69 months in 379 patients, averaging $1624 per patient. The margin (a ratio of hospital charges divided by hospital cost and a measure of profitability) was 54% higher at the bedside than the operating room and 9.6% higher than the radiology suite. Advances in the surgical techniques have been instrumental for the implementation of these bedside procedures.

Percutaneous tracheostomy can be performed by a dilatational technique that aims to create a skin incision and tracheal opening that is just large enough to accommodate a tracheostomy tube. The procedure is monitored bronchoscopically and its complication rate is similar to or lower than for open tracheostomy in the operating room. Percutaneous endoscopic gastrostomy is also performed under gastroscopic guidance and associated with excellent outcomes. Percutaneous vena cava filter placement is being increasingly performed at the bedside with a variety of techniques. Fluoroscopy is still the most widely practiced method, but it involves a C-arm, special ICU beds, ample space, and radiation concerns. External ultrasonography has also been used with success by certain groups, but visualization may be compromised by distended bowel. Intravascular ultrasonography is becoming a convenient and reliable technology that offers precise localization of the anatomical landmarks and prevents radiation exposure. Our group and others have been using this technique exclusively over the last 3 years, with excellent outcomes.

Hemodynamic monitoring is a crucial part of resuscitation. It is almost disheartening to admit that in an era of technological innovation most patients are still resuscitated according to no more accurate endpoints than a blood pressure and a heart rate. Although most authorities agree that these parameters are nonspecific and affected by a myriad of conditions, there is little progress in alternative methods. The pulmonary artery catheter is considered invasive and possibly dangerous. Arterial blood gases and lactate levels provide a good estimate of tissue perfusion but are episodic and irrelevant by the time the values are returned to the physician. Noninvasive methods that produce continuous and concurrent outflow of information are direly needed.

Cardiac bioimpedance has been explored as a noninvasive method to monitor cardiac output and found to be reliable in direct comparisons with the thermodilution technique. Heart rate variability has been targeted as a marker of the balance between the sympathetic and parasympathetic systems. A reduced heart rate variability, mentioned as “cardiac uncoupling,” independently predicts morbidity and mortality rates. Pulse pressure variability has been used similarly. It is suggested that it reflects volume status but confounded by arterial distensibility. Transcutaneous oxygen and carbon dioxide measurements can be performed very easily via a skin electrode and provide an estimate of skin tissue perfusion. Gastric tonometry and sublingual capnography have focused on the measurement of the partial pressure of carbon dioxide in the gastric and sublingual mucosa, respectively. Algorithms to estimate the pH are based on these measurements. Near-infrared spectroscopy emerges as a promising method that records the concentration of deoxygenated hemoglobin in superficial muscle beds. This is shown to reflect general organ tissue perfusion, correlate with resuscitation and outcome, and perform as well as arterial blood gases, while being noninvasive and continuous. Pulse contour analysis is a semi-invasive method because it relies on the presence of an arterial catheter. By measuring the aortic pressure waveform, it estimates the stroke volume and cardiac output but is being restricted by the need of frequent calibrations. Finally, transesophageal Doppler measurements have accurately guided resuscitation and provided information about cardiac filling, volume, and flow.

Resuscitation is fluid-based. The golden rule of resuscitating a hypotensive patient is to correct the blood pressure by administering intravenous fluids. This practice, although fundamentally correct, has led to indiscriminate administration of a large volume of fluids that resulted in aggravation of bleeding, development of compartment syndromes, and exacerbation of the inflammatory response. A new trend of hypovolemic resuscitation has gained momentum. After a prospective randomized trial that showed a restrictive fluid resuscitation policy to improve survival in patients with penetrating torso injuries, there is ongoing skepticism about crystalloid solutions as the ideal and abundant resuscitation medium. At the same time, the value of fresh frozen plasma (FFP) in the presence of severe bleeding is emphasized. Probably, lower packed red blood cells to FFP ratios than the traditional 3:1 and earlier FFP administration than the traditional release of FFP after the sixth unit of blood transfusion may produce better outcomes in the face of severe bleeding.

On the forefront of mechanical ventilation, lung-protective strategies and, in particular, ventilation by low tidal volumes have become the standard of care, following the Acute Respiratory Distress Syndrome Network (ARDSNet) trials. Higher levels of positive end-expiratory pressure (10 to 15 cm H₂O) are considered beneficial for patients with acute lung injury, if the plateau pressure does not rise to excessive points. Steroids have fallen out of favor for the treatment of adult respiratory distress syndrome (ARDS) because they produce no improvement in mortality rates and may actually worsen the mortality rate in specific subpopulations. Several innovative ventilatory strategies, such as high frequency oscillatory ventilation and airway pressure release ventilation, have been used in small studies and shown to produce unclear benefits but not yet been universally accepted.

Surgical infections remain a principal cause of morbidity and cost in the ICU. Surgical prophylaxis by a single, short-course, broad-spectrum antibiotic is as effective as longer and multiple antibiotic regimens. In randomized controlled studies and a meta-analysis there was no benefit of administering prophylactic antibiotics for more than 24 hours in the majority of surgical diseases. Selective decontamination has been practiced extensively in European ICUs but, despite favorable evidence, not been widely accepted in the United States. Empiric antibiotics should be administered early and in increased doses. Delay in the management of ventilator-associated pneumonia by empiric antibiotics leads to increased mortality rates. However, the physician should be prepared to de-escalate or even discontinue treatment as soon as data from cultures are available. Antibiotic cycling has prevented bacterial resistance, standardized empiric treatment, and reduced cost. Activated protein C was tested in the Recombinant Human Activated Protein C Worldwide Evaluation in Severe Sepsis study and found to reduce mortality among patients with sepsis and organ failures. At the same time it increased the risk of serious bleeding.

The debate about the ideal method of thromboprophylaxis continues. Unfractionated subcutaneous heparin is inferior to low-molecular-weight heparin and its use is called into question for critically ill surgical and trauma patients. Even low-molecular-weight heparin use is not founded on solid evidence, and there are several studies with contradicting results. Sequential compression devices may theoretically be an adequate method to encourage blood circulation through the extremities but in reality are subject to poor compliance by the patients. Inferior vena cava filters are designed to capture clots traveling from peripheral extremity veins to the lungs and in this way prevent pulmonary embolism. In a randomized study, the long-term incidence of pulmonary embolism and death from pulmonary embolism between those who received and those who did not receive a filter was not different. Those with a filter had a higher incidence of deep venous thrombosis. Removable filters are increasingly used, but the removal rates remain low. Selective inhibitors of factor Xa and thrombin have been developed, but there are no trials to document their effectiveness in critically ill populations.

The philosophy about nutrition in the ICU has changed drastically over the last 2 decades. From an era that favored total parenteral nutrition, intensivists have moved to an era of early enteral feeding. Parenteral nutrition is associated with severe complications and increased hospital length of stay. Several randomized studies and meta-analyses have documented a benefit of enteral over parenteral nutrition in terms of morbidity and infectious complications; a survival advantage was also found in specific subpopulations. Enteral nutrition should be instituted early but overfeeding according to old recommendations is no longer required and may be harmful. Immunonutrition presents a new topic, which is under intense scrutiny and debate. Immune-enhancing diets contain elements such as glutamine, arginine, nucleotides, or omega-3 fatty acids. In a stream of prospective randomized studies, these diets were found to decrease infectious morbidity and hospital length of stay. One multicenter study identified a benefit in survival. Newer evidence has shown that specific regimens produce no benefit and may actually be an unacceptable choice in septic patients due to an undesired increase of nitric oxide production. Different meta-analyses have produced different results, although the overall conclusion seems to be that immunonutrition improves outcomes in selected ICU populations.

Tight glycemic control by insulin has recently emerged as a therapeutic means of preventing infection. Following a prospective randomized study, many ICUs embarked on protocols of regulating glucose levels to levels below 110 mg/dL. However, new evidence suggests that the risk of hypoglycemia may be substantial and intensivists tend to relax the criteria of glucose control, although not returning to the previous practices.

Steroid deficiency in the ICU has been an insidious cause of persistent hypotension and inability to wean. A prospective randomized study showed that patients with adrenal insufficiency who received low-dose steroids have improved outcomes compared with those who did not. In the recently completed multicenter European Corticosteroid Therapy of Septic Shock (CORTICUS) study, steroids failed to show such benefit. Importantly, in the CORTICUS trial the corticotropin test did not reliably predict who would benefit from steroid treatment, and the measurement of serum cortisol levels did not help the decision making. However, it seems intuitively correct to administer low-dose steroids in patients with steroid insufficiency. Selecting these patients is difficult. The selection should not be based on measurement of cortisol levels but on clinical criteria (ie, failure to reverse shock despite adequate resuscitation) and in the absence of other correctable causes.

The science of ICU ethics remains scarce and physicians are frequently confronted with ethical dilemmas. The decision to discontinue care is probably one of the most frequent ones. Communication with families and legal representatives is key but should not be viewed as a vehicle for endless negotiations. With full respect to a family's views, a physician must have formed an independent medical opinion about the validity of continuing care or not. The physician should convey this decision effectively. Organ donation should always be part of the thought process when care is discontinued. The treating physician should not be the one making the donation request to the family. This role is better left to specially trained teams.

Innovation and discovery is abundant in the ICU. Principles change and more questions arise than answers. No one patient is the same. The intensivist is called to balance choices, concerns, and opportunities. The surgeon of the future will have a strong presence in the ICU, which in many ways will become an equally interesting and inviting environment as the operating room, satisfying the natural surgical curiosity, outcome-focused mentality, and need for change.