- Email: [email protected]
Gut dysfunction in critical illness
Clinical Nutrition (1997) 16:57-60 © Pearson Professional Ltd 1997
SUMMARY OF ROUND TABLE CONFERENCE Gut dysfunction in critical illness Chairmen: J. L. ROMBEAU and J. TAKALA
Introduction
plexity results from intricate interactions among various components such as the epithelium, genes used for protein synthesis, the microcapillary system and the gut-associated lymphoid tissue. Components of the intestinal epithelial barrier are both cellular and non-cellular. Cellular components include the enterocyte, goblet cell, intraepithelial lymphocyte, macrophage and entero-endocrine cell. Non-cellular components include the epithelial mucus and glycocalyx membrane. The most important component of the intestinal epithelial barrier is the enterocyte. This cell has both absorptive and barrier functions. An important anatomic attachment of enterocytes is the bridging of tight junctions, which include a key functional component known as the zonula occludens. Separation, widening or destruction of tight junctions provides the route for paracellular transmigration of bacteria and bacterial products through the epithelium into mesenteric lymph nodes and the portal circulation.
Little attention has been given to the status of the gastrointestinal tract in the critically ill patient. Traditional teachings in critical illness have promoted the dogma that the gut is dormant, metabolically inactive, and of little physiologic and pathologic significance. More recent information has refuted these long-standing beliefs. The intestinal tract indeed provides a number of important functions which in turn influence the clinical outcome of many critically ill patients. Its extensive absorptive area provides an important site for feeding and nutrient utilization. The gut also serves as a barrier to prevent abnormal absorption of intraluminal microbes and/or microbial products. Moreover, the gut is the most common source of nosocomial infections among critically ill patients. The intestinal tract is also the largest lymphoid organ in the body and it, in turn, plays an important role in the immune response of the stressed and critically ill patient. Gut dysfunction occurs frequently among critically ill patients. The severity, duration and mechanism of this dysfunction is often contingent upon the etiology of the critical illness. For example, the patient with hypovolemic shock and ensuing ischemia-reperfusion injury probably has a different mechanism of gut dysfunction from the patient with diffuse sepsis or polytrauma. A Round Table Conference on gut dysfunction in critical illness was recently conducted at the 16th International Symposium on Intensive Care and Emergency Medicine held in Brussels, Belgium. Physicians, surgeons and basic scientists from Europe and North America presented data and participated in discussions concerning this topic. This report summarizes the conclusions and recommendations emanating from this conference. A complete publication of the conference proceedings will be forthcoming (Intensive Care & Emergency Medicine Yearbook, Springer-Verlag). This report discusses the following with particular emphasis upon critical illness: (1) the normal structure and function of the intestinal barrier; (2) the pathogenesis of intestinal dysfunction; (3) monitoring intestinal function; and (4) potential treatments of gut dysfunction.
Molecular barrier Regulation of gene function is an important component of the intestinal barrier. Expression of several proto-oncogenes such as c-myc and c-jun, and increased mRNA levels for glucagon are associated with increased proliferation of enterocytes. The use of specific nutrients or growth factors to alter these gene products may have potential clinical significance.
Immune barrier The intestine is the largest reservoir of lymphocytes in the body. These cells reside in both the lamina propria and the intestinal epithelium. Importantly, intestinal lymphoid cells secrete many cytokines and mediators that stimulate and regulate the gut immune system. Based upon cell surface phenotype and functional capacity, the activation state of intestinal immune ceils is higher than that of similar cells in peripheral blood. Moreover, gut injury may result from an unchecked, inflammatory-mediated triggering of immune cells in the gut that are normally in an activated state and poised for effector function (i.e. cytokine production and cytotoxic function).
Normal intestinal barrier structure and function
Pathogenesis of intestinal dysfunction in critical illness
Epithelial barrier
There is no objective, clinically relevant definition of intestinal dysfunction in critical illness. Intolerance to an appropriate regimen of enteral nutrition is probably the
The epithelium is a key component of the intestinal barrier, and its structure and functions are complex. This corn57
58
SUMMARY OF ROUND TABLE CONFERENCE
most practical finding at the bedside. Despite obvious limitations to this definition, it provides a functional assessment with some clinical relevance. A more objective definition of intestinal dysfunction will undoubtedly emerge in the future and will most likely be based upon actual measurements of absorption and bacterial overgrowth. Ideally, any definition of intestinal dysfunction should also grade the severity of the dysfunction. The etiologic and pathogenetic mechanisms of intestinal dysfunction are multifactorial. Currently available data from clinical and animal studies are compatible with the following hypotheses: (1) intestinal bacteria are a major cause of complicating infections in hospitalized patients; (2) increased intestinal epithelial permeability may be a common factor facilitating bacterial translocation in high risk patients; and (3) hematogenous perfusion of different components of the gut wall may not meet the metabolic demands of the tissue. The metabolic demands of the gut wall vary because of functional heterogeneity. Under normal conditions, oxygen transport functions to maintain sufficient oxygen supply throughout the intestine. The regulation of tissue perfusion may be disrupted as the result of various insults. This may be the cause of hypoxic cell injury or ischemia-reperfusion injury; alternatively, ATP utilization may be suppressed and cell functions inhibited, especially in the mucosa. Hyperpermeability and bacterial translocation are distinct phenomena. Hyperpermeability is produced in experimental models by oxidant stress, ATP depletion, acidosis, nitric oxide and cytokine release. Inadequate tissue perfusion and endothelial damage may both contribute to and result from epithelial damage and hyperpermeability. It is likely that clinically evident intestinal dysfunction results from the interaction of these different mechanisms. The neutrophil appears to be an important mediator of gut dysfunction during the ischemia reperfusion injury. Although neutrophils contribute significantly to reperfusion injury, attention has been directed to extravascular cells, including mast cells. Recent findings have shown that mast ceils recruit and activate neutrophils, thereby potentiating reperfusion injury. These findings have therapeutic implications. Strategies designed to prevent mast cell degranulation may be efficacious in preventing an inappropriate inflammatory response. Modulating the function of mast cells by affecting nitric oxide production may be therapeutically beneficial. Local production of cytokines and oxygen radicals appear to mediate gut mucosal damage in catabolic states. During acute insult to the intestine, such as that which occurs during ischemia reperfusion (aortic clamping), the gut is an important cytokine producer, and the increased production of cytokines is prolonged even after a short ischemia/reperfusion event. Whether the increased cytokine production leads to SIRS or MOF is unknown. Additionally, the cause of this increased cytokine production is unclear. There are several possible initiators, such as enteric bacteria, endotoxin and ischemia. A resultant response from these multiple factors is most likely. Moreover, there appear
to be significant inter-individual differences among patients undergoing the same level of insult. Bacterial translocation is defined as the passage of bacteria or bacterial products from the intestine into mesenteric lymph nodes or the portal system. Bacterial translocation is due to hyperpermeability of the intestinal barrier.
Bacterial translocation models of gut dysfunction in experimental and in selected clinical conditions While the importance of bacterial translocation to multiple organ damage is clear in the experimental setting, the relevance to critically ill patients is uncertain. Moreover, investigations in patients undergoing major abdominal trauma have failed to confirm the frequency of bacterial translocation as an important cause of sepsis. Additionally, septic shock rarely produces bacterial translocation even in patients who are destined to die. It is unclear whether bacterial translocation is a mechanistic cause or the result of late MOF in critically ill patients.
Monitoring of gut function Due to the lack of objective, uniform definitions of gut dysfunction, monitoring of gut function must be based on indirect indicators. Tolerance to enteral feeding is probably the most commonly used indicator in the clinical setting. Its relevance can be improved when performed in the context of a predefined feeding protocol. Gastrointestinal bleeding has been used as an indicator of intestinal failure, especially in organ failure scoring. While it is certainly a sign of gastrointestinal organ damage, it does not quantitatively assess dysfunction, and is not suitable for monitoring due to its variable characteristics. Absorption tests and markers of increased permeability assess relevant aspects of gut function, and they can be relatively easily applied for research purposes. With the current methodology, absorption and permeability tests are not suitable for clinical monitoring in the intensive care setting. Gastrointestinal tonometry provides information of the adequacy of splanchnic perfusion. While it does not reflect gut function directly, it is currently the most promising technique for the clinical monitoring of the gut in the intensive care setting. Gastric mucosal acidosis is associated with increased mortality in various groups of intensive care patients. Therapy aimed at preventing and correcting gastric mucosal acidosis may improve outcome, but the role of tonometry in monitoring the response to therapeutic interventions has not been well established. The methodology is susceptible to technical and operator-related variability and errors, and meticulous care is required to assure good reproducibility. As a local measurement, its sensitivity to concurrent perfusion changes in different parts of the gastrointestinal tract is not known. The recent introduction of gas tonornetry using an automated sampling of gastric gas CO 2 is a promising new modification of gastrointestinal tonometry.
CLINICAL NUTRITION
Potential therapies Three main strategies can be used to prevent gut dysfunction and to treat existing functional abnormalities: (1) improvement of tissue perfusion in order to balance tissue oxygen supply and blood flow to the metabolic demands; (2) prevention of bacterial overgrowth to selective decontamination; and (3) support of the structural and functional intestinal integrity by enteral feeding.
Enhancement of intestinal perfusion Regulation of gut perfusion in the clinical intensive care setting is poorly understood. Distribution of blood flow within the intestinal wall is heterogeneous, and attempts to improve perfusion may have variable effects in the different layers of the gut. The mucosa, with its high metabolic demands and villous microvascular anatomy, is particularly susceptible to inadequate perfusion. Hypovolemia reduces gastrointestinal blood flow, and splanchnic vasoconstriction is prolonged after correction of hypovolemia. Accordingly, prevention and aggressive treatment of hypovolemia is the cornerstone of supporting gut perfusion. The effects of adrenergic and other vasoactive agents on gut perfusion are variable and are insufficiently documented. Data from experimental studies and other patient groups cannot be extrapolated to the intensive care patients. There is increasing evidence that the underlying clinical condition may markedly modify the response to commonly used vasoactive agents. Furthermore, the splanchnic blood flow responses may vary markedly between individuals. In general, therapeutic interventions that increase cardiac output are also likely to increase total splanchnic blood flow in patients with relatively intact vasoregulation. In contrast, in sepsis and SIRS, the relationship between changes in cardiac output and splanchnic blood flow are less consistent. Vasoactive drugs may redistribute tissue perfusion within the splanchnic region, and perfusion may be impaired locally despite increased total regional flow. Attempts to improve perfusion may also result in metabolic alterations, and modify the balance between oxygen supply and demand. Ideally, gut perfusion should be monitored together with functional and metabolic markers. Due to the heterogeneity of gut perfusion and its responses to therapeutic interventions, future therapies should be tailored for individual patients.
Selective digestive decontamination Loss of bacterial homeostasis is a characteristic feature of gut dysfunction in critically ill patients. Selective decontamination of the gastrointestinal tract reduces nosocomial infections. Effects on mortality are controversial and are related at least in part to different decontamination regimens (efficiency of decontamination) and patient selection. Better selection of target populations, compliance with strict regimens, and monitoring the efficiency of decontamination
59
may help to extend the effects on nosocomial infection to enhanced recovery.
Nutritional-metabolic treatment Nutritional support, in the form of parenteral and enteral nutrition, is now a mandatory component of the care of most critically ill patients. To maintain nutritional status and to prevent ensuing gut dysfunction, critically ill or severely injured patients are now fed early in their disease course. The following indications have been proposed as guidelines for the initiation of nutritional support in critically ill or severely injured patients: absence of nutrition for 5-7 days, anticipated duration of illness more than 10 days, and presence of malnutrition as documented by a loss of weight greater than 10%. The most important stimulus for intestinal epithelial growth and function is the presence of nutrients within the gut lumen. These nutrients mediate gut growth, and function both directly and indirectly. The direct presence of nutrients provides local substrate for cellular oxidation and creates a mechanical stimulus for increased proliferation of enterocytes. The indirect trophic effects of enteral nutrients are mediated through the increased production of trophic gastrointestinal hormones which act via autocrine, paracrine and endocrine pathways. If the gastrointestinal tract can be used safely, it is the preferred route for nutrient delivery. A review of the prospective, randomized controlled trials of the use of enteral nutrition in critically ill patients demonstrates that early enteral nutrition is safe, well tolerated and reduces postoperative sepsis in trauma patients. Whether similar benefits occur in critically ill patients is less clear because of the lack of well controlled studies. Early enteral nutrition should be encouraged in most critically ill patients within the context of a careful protocol for nutritional delivery and monitoring. Ideal formulae and volumes have yet to be determined. Most authorities in this field believe that feeding into the jejunum is better tolerated than delivery of nutrients into the stomach. Recently developed liquid formula diets have been supplemented with exogenous nutrients and intestinal fuels such as glutamine, arginine, fish oils and nucleotides (immune-enhancing diets). Diminution of glutamine availability leads to increased intestinal permeability 'which can be counteracted by enriching parenteral nutritional regimens with glutamine dipeptides. The depleted .gut is characterized by an inflammatory response in which villus atrophy coincides with increased crypt cell proliferation. Glutamine appears to have an anti-inflammatory effect in this setting. Unfortunately, there are no published controlled trials of critically ill patients to evaluate the effects of e.nteral diets supplemented solely with glutamine. Recent trials in postoperative cancer and trauma patients have shown that the immune-enhancing diets improve clinical outcome and reduce sepsis. The mechanisms of these effects are unclear. Growth factors exert trophic effects on the gut and promote intestinal cell proliferation. Proliferative effects are less important in the critically ill patient than in the
60
SUMMARY OF ROUND TABLE CONFERENCE
patient with postoperative intestinal resection. Currently, there are not sufficient well controlled clinical data to justify the use of expensive recombinant produced growth factors for critically ill patients.
Summary and conclusions There is recent evidence that intestinal function is an important determinant in the outcome of critically ill patients. The barrier function is an important characteristic of the gut. Its workings are complex and it consists of epithelial, molecular and immune components. The pathogenesis of gut dysfunction among critically ill patients is multifactorial, consisting of the quality and quantity of micro-organisms, the permeability of the epithelium and the extent of the vascular perfusion. A practical bedside measure of gut dysfunction is intolerance to enteral feedings. Potential therapies for gut dysfunction include the use of vasoactive drugs to enhance perfusion, selective decontamination to decrease bacteria and endotoxin, and the early and frequent administration of enteral nutrients.
List of participants Carlet J (France) Carpentier (Belgium)
Evans T W (UK) Fink M P (USA) Furst P (Germany) Goris R J A (The Netherlands) Groeneveld J A B (The Netherlands) Haglund (Sweden) Kubes P (Canada) Marshall J C (Canada) McVay L (USA) Moore F A (USA) Mythen M (USA) Payen D M (France) Radermacher P (Germany) Redl H (Austria) Rombean J L (USA) Schumacker P (USA) Singer M (UK) Soeters P (The Netherlands) Stoutenbeck C (The Netherlands) Takala J (Finland) Vallet N (France) Vincent J L (Belgium) Wells C L (USA) Zhang H (Belgium) Ziegler T R (USA)
SUMMARY OF ROUND TABLE CONFERENCE Gut dysfunction in critical illness Chairmen: J. L. ROMBEAU and J. TAKALA
Introduction
plexity results from intricate interactions among various components such as the epithelium, genes used for protein synthesis, the microcapillary system and the gut-associated lymphoid tissue. Components of the intestinal epithelial barrier are both cellular and non-cellular. Cellular components include the enterocyte, goblet cell, intraepithelial lymphocyte, macrophage and entero-endocrine cell. Non-cellular components include the epithelial mucus and glycocalyx membrane. The most important component of the intestinal epithelial barrier is the enterocyte. This cell has both absorptive and barrier functions. An important anatomic attachment of enterocytes is the bridging of tight junctions, which include a key functional component known as the zonula occludens. Separation, widening or destruction of tight junctions provides the route for paracellular transmigration of bacteria and bacterial products through the epithelium into mesenteric lymph nodes and the portal circulation.
Little attention has been given to the status of the gastrointestinal tract in the critically ill patient. Traditional teachings in critical illness have promoted the dogma that the gut is dormant, metabolically inactive, and of little physiologic and pathologic significance. More recent information has refuted these long-standing beliefs. The intestinal tract indeed provides a number of important functions which in turn influence the clinical outcome of many critically ill patients. Its extensive absorptive area provides an important site for feeding and nutrient utilization. The gut also serves as a barrier to prevent abnormal absorption of intraluminal microbes and/or microbial products. Moreover, the gut is the most common source of nosocomial infections among critically ill patients. The intestinal tract is also the largest lymphoid organ in the body and it, in turn, plays an important role in the immune response of the stressed and critically ill patient. Gut dysfunction occurs frequently among critically ill patients. The severity, duration and mechanism of this dysfunction is often contingent upon the etiology of the critical illness. For example, the patient with hypovolemic shock and ensuing ischemia-reperfusion injury probably has a different mechanism of gut dysfunction from the patient with diffuse sepsis or polytrauma. A Round Table Conference on gut dysfunction in critical illness was recently conducted at the 16th International Symposium on Intensive Care and Emergency Medicine held in Brussels, Belgium. Physicians, surgeons and basic scientists from Europe and North America presented data and participated in discussions concerning this topic. This report summarizes the conclusions and recommendations emanating from this conference. A complete publication of the conference proceedings will be forthcoming (Intensive Care & Emergency Medicine Yearbook, Springer-Verlag). This report discusses the following with particular emphasis upon critical illness: (1) the normal structure and function of the intestinal barrier; (2) the pathogenesis of intestinal dysfunction; (3) monitoring intestinal function; and (4) potential treatments of gut dysfunction.
Molecular barrier Regulation of gene function is an important component of the intestinal barrier. Expression of several proto-oncogenes such as c-myc and c-jun, and increased mRNA levels for glucagon are associated with increased proliferation of enterocytes. The use of specific nutrients or growth factors to alter these gene products may have potential clinical significance.
Immune barrier The intestine is the largest reservoir of lymphocytes in the body. These cells reside in both the lamina propria and the intestinal epithelium. Importantly, intestinal lymphoid cells secrete many cytokines and mediators that stimulate and regulate the gut immune system. Based upon cell surface phenotype and functional capacity, the activation state of intestinal immune ceils is higher than that of similar cells in peripheral blood. Moreover, gut injury may result from an unchecked, inflammatory-mediated triggering of immune cells in the gut that are normally in an activated state and poised for effector function (i.e. cytokine production and cytotoxic function).
Normal intestinal barrier structure and function
Pathogenesis of intestinal dysfunction in critical illness
Epithelial barrier
There is no objective, clinically relevant definition of intestinal dysfunction in critical illness. Intolerance to an appropriate regimen of enteral nutrition is probably the
The epithelium is a key component of the intestinal barrier, and its structure and functions are complex. This corn57
58
SUMMARY OF ROUND TABLE CONFERENCE
most practical finding at the bedside. Despite obvious limitations to this definition, it provides a functional assessment with some clinical relevance. A more objective definition of intestinal dysfunction will undoubtedly emerge in the future and will most likely be based upon actual measurements of absorption and bacterial overgrowth. Ideally, any definition of intestinal dysfunction should also grade the severity of the dysfunction. The etiologic and pathogenetic mechanisms of intestinal dysfunction are multifactorial. Currently available data from clinical and animal studies are compatible with the following hypotheses: (1) intestinal bacteria are a major cause of complicating infections in hospitalized patients; (2) increased intestinal epithelial permeability may be a common factor facilitating bacterial translocation in high risk patients; and (3) hematogenous perfusion of different components of the gut wall may not meet the metabolic demands of the tissue. The metabolic demands of the gut wall vary because of functional heterogeneity. Under normal conditions, oxygen transport functions to maintain sufficient oxygen supply throughout the intestine. The regulation of tissue perfusion may be disrupted as the result of various insults. This may be the cause of hypoxic cell injury or ischemia-reperfusion injury; alternatively, ATP utilization may be suppressed and cell functions inhibited, especially in the mucosa. Hyperpermeability and bacterial translocation are distinct phenomena. Hyperpermeability is produced in experimental models by oxidant stress, ATP depletion, acidosis, nitric oxide and cytokine release. Inadequate tissue perfusion and endothelial damage may both contribute to and result from epithelial damage and hyperpermeability. It is likely that clinically evident intestinal dysfunction results from the interaction of these different mechanisms. The neutrophil appears to be an important mediator of gut dysfunction during the ischemia reperfusion injury. Although neutrophils contribute significantly to reperfusion injury, attention has been directed to extravascular cells, including mast cells. Recent findings have shown that mast ceils recruit and activate neutrophils, thereby potentiating reperfusion injury. These findings have therapeutic implications. Strategies designed to prevent mast cell degranulation may be efficacious in preventing an inappropriate inflammatory response. Modulating the function of mast cells by affecting nitric oxide production may be therapeutically beneficial. Local production of cytokines and oxygen radicals appear to mediate gut mucosal damage in catabolic states. During acute insult to the intestine, such as that which occurs during ischemia reperfusion (aortic clamping), the gut is an important cytokine producer, and the increased production of cytokines is prolonged even after a short ischemia/reperfusion event. Whether the increased cytokine production leads to SIRS or MOF is unknown. Additionally, the cause of this increased cytokine production is unclear. There are several possible initiators, such as enteric bacteria, endotoxin and ischemia. A resultant response from these multiple factors is most likely. Moreover, there appear
to be significant inter-individual differences among patients undergoing the same level of insult. Bacterial translocation is defined as the passage of bacteria or bacterial products from the intestine into mesenteric lymph nodes or the portal system. Bacterial translocation is due to hyperpermeability of the intestinal barrier.
Bacterial translocation models of gut dysfunction in experimental and in selected clinical conditions While the importance of bacterial translocation to multiple organ damage is clear in the experimental setting, the relevance to critically ill patients is uncertain. Moreover, investigations in patients undergoing major abdominal trauma have failed to confirm the frequency of bacterial translocation as an important cause of sepsis. Additionally, septic shock rarely produces bacterial translocation even in patients who are destined to die. It is unclear whether bacterial translocation is a mechanistic cause or the result of late MOF in critically ill patients.
Monitoring of gut function Due to the lack of objective, uniform definitions of gut dysfunction, monitoring of gut function must be based on indirect indicators. Tolerance to enteral feeding is probably the most commonly used indicator in the clinical setting. Its relevance can be improved when performed in the context of a predefined feeding protocol. Gastrointestinal bleeding has been used as an indicator of intestinal failure, especially in organ failure scoring. While it is certainly a sign of gastrointestinal organ damage, it does not quantitatively assess dysfunction, and is not suitable for monitoring due to its variable characteristics. Absorption tests and markers of increased permeability assess relevant aspects of gut function, and they can be relatively easily applied for research purposes. With the current methodology, absorption and permeability tests are not suitable for clinical monitoring in the intensive care setting. Gastrointestinal tonometry provides information of the adequacy of splanchnic perfusion. While it does not reflect gut function directly, it is currently the most promising technique for the clinical monitoring of the gut in the intensive care setting. Gastric mucosal acidosis is associated with increased mortality in various groups of intensive care patients. Therapy aimed at preventing and correcting gastric mucosal acidosis may improve outcome, but the role of tonometry in monitoring the response to therapeutic interventions has not been well established. The methodology is susceptible to technical and operator-related variability and errors, and meticulous care is required to assure good reproducibility. As a local measurement, its sensitivity to concurrent perfusion changes in different parts of the gastrointestinal tract is not known. The recent introduction of gas tonornetry using an automated sampling of gastric gas CO 2 is a promising new modification of gastrointestinal tonometry.
CLINICAL NUTRITION
Potential therapies Three main strategies can be used to prevent gut dysfunction and to treat existing functional abnormalities: (1) improvement of tissue perfusion in order to balance tissue oxygen supply and blood flow to the metabolic demands; (2) prevention of bacterial overgrowth to selective decontamination; and (3) support of the structural and functional intestinal integrity by enteral feeding.
Enhancement of intestinal perfusion Regulation of gut perfusion in the clinical intensive care setting is poorly understood. Distribution of blood flow within the intestinal wall is heterogeneous, and attempts to improve perfusion may have variable effects in the different layers of the gut. The mucosa, with its high metabolic demands and villous microvascular anatomy, is particularly susceptible to inadequate perfusion. Hypovolemia reduces gastrointestinal blood flow, and splanchnic vasoconstriction is prolonged after correction of hypovolemia. Accordingly, prevention and aggressive treatment of hypovolemia is the cornerstone of supporting gut perfusion. The effects of adrenergic and other vasoactive agents on gut perfusion are variable and are insufficiently documented. Data from experimental studies and other patient groups cannot be extrapolated to the intensive care patients. There is increasing evidence that the underlying clinical condition may markedly modify the response to commonly used vasoactive agents. Furthermore, the splanchnic blood flow responses may vary markedly between individuals. In general, therapeutic interventions that increase cardiac output are also likely to increase total splanchnic blood flow in patients with relatively intact vasoregulation. In contrast, in sepsis and SIRS, the relationship between changes in cardiac output and splanchnic blood flow are less consistent. Vasoactive drugs may redistribute tissue perfusion within the splanchnic region, and perfusion may be impaired locally despite increased total regional flow. Attempts to improve perfusion may also result in metabolic alterations, and modify the balance between oxygen supply and demand. Ideally, gut perfusion should be monitored together with functional and metabolic markers. Due to the heterogeneity of gut perfusion and its responses to therapeutic interventions, future therapies should be tailored for individual patients.
Selective digestive decontamination Loss of bacterial homeostasis is a characteristic feature of gut dysfunction in critically ill patients. Selective decontamination of the gastrointestinal tract reduces nosocomial infections. Effects on mortality are controversial and are related at least in part to different decontamination regimens (efficiency of decontamination) and patient selection. Better selection of target populations, compliance with strict regimens, and monitoring the efficiency of decontamination
59
may help to extend the effects on nosocomial infection to enhanced recovery.
Nutritional-metabolic treatment Nutritional support, in the form of parenteral and enteral nutrition, is now a mandatory component of the care of most critically ill patients. To maintain nutritional status and to prevent ensuing gut dysfunction, critically ill or severely injured patients are now fed early in their disease course. The following indications have been proposed as guidelines for the initiation of nutritional support in critically ill or severely injured patients: absence of nutrition for 5-7 days, anticipated duration of illness more than 10 days, and presence of malnutrition as documented by a loss of weight greater than 10%. The most important stimulus for intestinal epithelial growth and function is the presence of nutrients within the gut lumen. These nutrients mediate gut growth, and function both directly and indirectly. The direct presence of nutrients provides local substrate for cellular oxidation and creates a mechanical stimulus for increased proliferation of enterocytes. The indirect trophic effects of enteral nutrients are mediated through the increased production of trophic gastrointestinal hormones which act via autocrine, paracrine and endocrine pathways. If the gastrointestinal tract can be used safely, it is the preferred route for nutrient delivery. A review of the prospective, randomized controlled trials of the use of enteral nutrition in critically ill patients demonstrates that early enteral nutrition is safe, well tolerated and reduces postoperative sepsis in trauma patients. Whether similar benefits occur in critically ill patients is less clear because of the lack of well controlled studies. Early enteral nutrition should be encouraged in most critically ill patients within the context of a careful protocol for nutritional delivery and monitoring. Ideal formulae and volumes have yet to be determined. Most authorities in this field believe that feeding into the jejunum is better tolerated than delivery of nutrients into the stomach. Recently developed liquid formula diets have been supplemented with exogenous nutrients and intestinal fuels such as glutamine, arginine, fish oils and nucleotides (immune-enhancing diets). Diminution of glutamine availability leads to increased intestinal permeability 'which can be counteracted by enriching parenteral nutritional regimens with glutamine dipeptides. The depleted .gut is characterized by an inflammatory response in which villus atrophy coincides with increased crypt cell proliferation. Glutamine appears to have an anti-inflammatory effect in this setting. Unfortunately, there are no published controlled trials of critically ill patients to evaluate the effects of e.nteral diets supplemented solely with glutamine. Recent trials in postoperative cancer and trauma patients have shown that the immune-enhancing diets improve clinical outcome and reduce sepsis. The mechanisms of these effects are unclear. Growth factors exert trophic effects on the gut and promote intestinal cell proliferation. Proliferative effects are less important in the critically ill patient than in the
60
SUMMARY OF ROUND TABLE CONFERENCE
patient with postoperative intestinal resection. Currently, there are not sufficient well controlled clinical data to justify the use of expensive recombinant produced growth factors for critically ill patients.
Summary and conclusions There is recent evidence that intestinal function is an important determinant in the outcome of critically ill patients. The barrier function is an important characteristic of the gut. Its workings are complex and it consists of epithelial, molecular and immune components. The pathogenesis of gut dysfunction among critically ill patients is multifactorial, consisting of the quality and quantity of micro-organisms, the permeability of the epithelium and the extent of the vascular perfusion. A practical bedside measure of gut dysfunction is intolerance to enteral feedings. Potential therapies for gut dysfunction include the use of vasoactive drugs to enhance perfusion, selective decontamination to decrease bacteria and endotoxin, and the early and frequent administration of enteral nutrients.
List of participants Carlet J (France) Carpentier (Belgium)
Evans T W (UK) Fink M P (USA) Furst P (Germany) Goris R J A (The Netherlands) Groeneveld J A B (The Netherlands) Haglund (Sweden) Kubes P (Canada) Marshall J C (Canada) McVay L (USA) Moore F A (USA) Mythen M (USA) Payen D M (France) Radermacher P (Germany) Redl H (Austria) Rombean J L (USA) Schumacker P (USA) Singer M (UK) Soeters P (The Netherlands) Stoutenbeck C (The Netherlands) Takala J (Finland) Vallet N (France) Vincent J L (Belgium) Wells C L (USA) Zhang H (Belgium) Ziegler T R (USA)