sign in sign up
<<
Home , Tachypnea

2 DIAGNOSIS AND SOURCE OF : THE UTILITY OF CLINICAL FINDINGS

Kenneth Cruz R. Phillip Dellinger

Sepsis continues to be one of the leading causes of mortality in hospitals today [1,2]. While the advent of intensive care units (ICUs) equipped with sophisticated monitoring systems and specially trained critical care physicians has likely reduced the mortality associated with this crippling syndrome, sepsis still remains an impressive medical foe. Therefore, early recognition of this clinical syndrome prompting empiric medical therapy likely improves patient outcome. Many signs, symptoms, and laboratory abnormalities may suggest the presence of sepsis, although their sensitivity and specificity vary considerably. History, first and foremost, may suggest a source of infection as well as pre-existing breaches in the immune system that put the patient at risk. Identifying the immunocompromised state intensifies the search for subtle signs of sepsis. Patients with compromised immune systems include individuals at the extremes of age, the neonatal, and the geriatric populations. Other common immunocompromised states including the human immunodeficiency virus (HIV), chemotherapy for cancer, post-transplant; chronic renal and liver failure patients also have suboptimal immune systems. There are also a number of uncommon congenital immunodeficiency syndromes that predispose individuals to infection. Our integumentary system maintains a vital barrier to infection, but, when compromised, either through severe, widespread burns, or by iatrogenic interventions such as central venous catheters, may hide subtle sources of infection. The history can also provide clues to the type of causative organism allowing empiric antibiotic therapy that targets likely organisms, potentially improving outcomes. Recent hospitalization or long-term care suggests the 12 Diagnosis and Source of Sepsis

possibility of virulent Gram-negative organisms. Gram-positive organisms should be entertained by the presence of recently implanted prosthetic devices such as heart valves, joints, pacemakers/defibrillators, and indwelling catheters (urinary or vascular). Fungal infection is more likely in patients with T-cell dysfunction, long term steroid use, and broad-spectrum antibiotic use.

A complete review of system symptomology can procure diagnostic clues to the source of infection (Table 1). These include, of course, general clues such as fever and chills, which suggest bacteremia or endotoxemia. Central nervous system (CNS) symptoms include headache, dizziness, and visual changes. Eye pain and discharge, ear pain and discharge, nasal discharge that is either purulent or bloody implies possible upper respiratory tract infection. , especially with thickened or colored , sore throat, dyspnea, Sepsis 13

/, can suggest a cardiorespiratory infection. Nausea/vomiting, diarrhea, jaundice, abdominal pain or distention, may suggest an abdominal source. Dysuria or hematuria indicates possible urinary tract infection. Also, dermatologic clues such as rashes, bullae, vesicles, and their distribution is important. Patient symptoms allow a more directed physical examination. History and when combined with physical examination and laboratory findings may point the way to the radiographic identification of niduses of infection that require source control (Figures 1 and 2). 14 Diagnosis and Source of Sepsis

VITAL SIGNS

In all patients, vital signs are the initial focus of the evaluation.

Fever

One of the cardinal signs associated with sepsis is fever. Fever is defined as a temperature greater then 38 °C. may also suggest sepsis. Temperature is usually obtained with an oral thermometer. However, in the hyperventilating patient, this method may be misleading and underestimate the actual core temperature [3]. When infection is suspected, a core temperature, such as rectal, should be obtained to more accurately assess the presence of hyper- or hypothermia. Bacterial products trigger the production of endogenous cytokines [1]. Tumor necrosis factor (TNF) and interleukin (IL)-l are pyrogens known to be released by macrophages during sepsis [4], which, through the production of prostaglandin and cyclic AMP, increase the baseline temperature by interacting with the thermoregulatory center of the hypothalamus [5]. Animal models suggest that fever protects the host species in sepsis. Classic studies Sepsis 15

with poikilotherms in the past have suggested the benefits of during sepsis. In these experiments, cold-blooded animals were placed in separate environments during infection – with a heat source or without. Animals that moved closer to the heat source had reduced mortality most likely because they were able to elevate their body temperature [6]. In addition, salicylate-induced antipyresis has been linked to increased mortality in animal models. These studies, while simple, suggest the benefits of fever, or hyperthermia, during sepsis and how elevated temperatures may serve as a physiologic defense mechanism. The reported incidence of hypothermia in sepsis varies. In a study by LeGall and colleagues, hypothermia, defined as a temperature less than 36 °C, was found in 15% of intensive care unit patients [8]. However, in another multi-institutional study by Clemmer et al, hypothermia, defined in their study as a temperature <35.5 °C, occurred in only 9% of cases of sepsis. Summarizing the literature as it relates to hypothermia is difficult due to a lack of a standard definition. Prior studies have linked hypothermia to poor prognosis in Gram-negative septicemia [9]. Detrimental physiologic effects of hypothermia include decreased immune function. At the onset of sepsis, hypothermic patients have greater physiologic derangements when compared with febrile patients, including central nervous dysfunction and circulatory shock. Renal function and need for were reported to be similar in both groups [10]. Hypothermia, therefore, identifies a group of septic patients with more severe physiologic abnormalities and increased mortality.

Respiratory Rate

While fever remains the most frequent diagnostic clue for sepsis, tachypnea is commonly exhibited in the early stages of sepsis [11]. has multiple possible explanations. Sepsis may directly induce central hyperventilation and respiratory alkalemia by a mechanism that is not clearly defined. Another mechanism for tachypnea is as compensation for metabolic (lactic) acidosis. Lactate production may be related to oxygen debt, or other metabolic abnormalities such as abnormal glycolysis related to liver dysfunction in sepsis [12]. Hypotension associated with sepsis has also been postulated as an explanation for tachypnea. Sepsis-induced hypotension can potentially produce acute respiratory alkalosis through decreased blood flow to aortic and carotid baroreceptors. Decreased pressure at these sites triggers the body to increase its in order to compensate. This has also been 16 Diagnosis and Source of Sepsis

hypothesized to account for the similar responses seen in cardiogenic and [13]. As the mechanistic theory of sepsis has evolved, the elucidation of cytokine-mediated damage has become clearer. Cytokines, such as TNF and IL-1, may be responsible for the tachypnea observed during sepsis through direct stimulation of the brainstem. There is, however, a lack of literature support for this theory. -induced tachypnea provides another possible physiologic explanation for the respiratory alkalosis observed in sepsis. Early in sepsis, ventilation/perfusion mismatching in the absence of any evidence of acute lung injury (ALI) might induce hypoxemia and associated tachypnea. ALI could also be associated with tachypnea, both as a response to hypoxemia and due to increased interstitial water stimulation of J-receptors.

Tachycardia and Hypotension

Hemodynamic derangements are commonly associated with sepsis. Tachycardia is an early clue to the diagnosis of sepsis and, as in exercise, is triggered to increase cardiac output. The average initial heart rate in septic individuals was approximately 120 beats per minute (bpm) in one study. In septic patients with tachycardia, a reduction below 106 bpm within the first twenty-four hours predicted improved survival [14]. Tachycardia is also observed despite the absence of hypotension. Numerous mediators have been implicated in the pathogenesis of hypotension during sepsis. Throughout the years, a significant amount of research has been dedicated to finding one single culprit. However, as more mediators are identified, the presumption that one single endogenous mediator is primarily responsible for the pathogenesis of shock becomes less likely. Endogenous substances implicated in the endothelial dysfunction associated with sepsis include predominantly cytokines, nitric oxide (NO, previously referred to as endothelial derived relaxation factor [EDRF]), arachidonic acid metabolites, neutrophils and their interactions with endothelial cells, opioids, complement, and bradykinins [15]. The physiologic actions of these substances will be explained in more detail later in this text. Septic hemodynamic parameters are described as ‘hyperdynamic’, alluding to the increased cardiac index (CI) and reduced systemic vascular resistance (SVR) [16]. Severe sepsis may be associated with a reduction in systemic vascular resistance (SVR) through the induction of NO synthase (NOS). This reduction in SVR initiates compensatory increases in heart rate Sepsis 17

to maintain adequate blood pressure. The following equations demonstrate the mechanisms involved:

SVR x cardiac output = mean arterial pressure (MAP) cardiac output = stroke volume (SV) x heart rate (HR)

Therefore, the body attempts to compensate for the reduction in SVR by increasing cardiac output. Compensatory mechanisms available in the septic patient to maintain adequate cardiac output include: 1) increasing heart rates; and 2) in the presence of decreased ejection fraction, increasing left end diastolic volumes. In severe sepsis, myocardial function is depressed and left ventricular preload is compromised secondary to capillary leak and venodilation. Therefore, increasing left ventricular end diastolic volumes and increasing heart rate is necessary to maintain adequate blood pressure [17]. However, in time, as the mediators of sepsis and inflammation progress, the reduction in SVR may outpace the ability of the body to compensate. Therefore hypotension, and eventually shock, ensue. Shock is defined as vasopressor requirements to maintain blood pressure and compromised tissue perfusion leading to cellular dysfunction.

SYSTEMIC MANIFESTATIONS OF SEPSIS (TABLE 2)

Central Nervous System Findings

Encephalopathy, a common sepsis-induced CNS presentation, is defined as any mental dysfunction due to the systemic manifestations of sepsis. Manifestations of this disorder include lethargy, somnolence, agitation, disorientation, confusion, and at the extreme, obtundation. Sepsis has been demonstrated to produce brain dysfunction [18]. Patients with acute, sepsis- induced altered mental status have increased mortality when compared to those with normal mental capabilities or prior mental status changes [19]. Although altered mental status has been used as a criterion for enrolment in sepsis clinical trials, it is often difficult to define a constant threshold as well as to separate out altered mental status due to other causes. Several theories have been proposed to explain the CNS abnormalities of sepsis. Causes may be direct (circulating toxins and inflammatory mediators) and indirect (hypotension and hypoglycemia). Abnormal ammo acid metabolism is known to be a cause of encephalopathy as well as abnormal amino acid transport across the blood-brain barrier [20]. This cause of altered mental status is more related to hepatic dysfunction than to CNS dysfunction. Altered cerebral perfusion has also been implicated in the pathophysiology of 18 Diagnosis and Source of Sepsis

CNS dysfunction in sepsis. Other theories implicate a breakdown of the blood-brain barrier, which allows the crossing of inflammatory cytokines as well as other humoral mediators into the CNS. The migration into the CNS of these inflammatory mediators, which affect astrocytes and neurons, causes encephalopathy irrespective of other confounding factors [21]. It remains unclear whether encephalopathy associated with sepsis contributes to the mortality or does the encephalopathy merely indicate the presence of severe global organ dysfunction? Sepsis 19

Cardiopulmonary Findings

As previously mentioned, the hemodynamic findings during sepsis include a reduced SVR and an elevated CI. The compensatory increase in cardiac output functions to maintain an adequate blood pressure despite increased oxygen extraction. However, should the physiologic abnormalities associated with sepsis continue, overwhelming the cardiovascular compensatory mechanisms initiated, then hypotension and shock ensue. due to increased lactate production and decreased clearance is one of the hallmark manifestations of severe sepsis and . The overproduction of lactate created by global organ hypoperfusion (distributive shock) was initially thought to be the primary cause for acidosis with ensuing anaerobic glycolysis, producing energy in the absence of oxygen with the by-product being lactate [22]. Several additional theories regarding the level of cellular dysfunction have arisen to explain the metabolic abnormalities of sepsis. Hypotheses regarding dysfunction at the regional level, microcirculatory level, cellular metabolic level, and the level of liver metabolism have arisen. The ratio, which can be measured indirectly using the lactate/pyruvate ratio, is a marker of global hypoperfusion. The lactate/pyruvate (L/P=) ratio increases (>10) during cellular and coincides with a poor prognosis [23]. The presence of circulatory dysfunction, demonstrated at the gastrointestinal level with tonometry, has been used to indicate regional hypoperfusion, allowing the translocation of endotoxin and bacteria that fuel sepsis. Low intramucosal pH has been shown to correlate with systemic hypoperfusion and with increased morbidity and mortality [24]. Alternatively, microcirculatory changes leading to cellular dysoxia may underlie the hemodynamic and metabolic changes seen during hypoperfusion in sepsis [25]. However, there has been some question raised as to whether elevated lactate levels correlate with oxygen debt, or rather suggest a metabolic abnormality. Recently, there has been some interest in the hypothesis that hyperlactatemia may not symbolize global organ hypoperfusion during sepsis. Instead, metabolic abnormalities may produce such levels. There has been some suggestion in the past that pyruvate dehydrogenase dysfunction in sepsis may lead to hyperlactatemia [26]. Another hypothesis argues that lactate production by an adrenergically stimulated increases despite evidence of normal oxygen delivery and consumption by skeletal muscles [27]. Other studies have shown that despite increased oxygen consumption, pyruvate oxidation, and lactate production during sepsis, the production of excess pyruvate accounts for the hyperlactatemia seen [28]. Findings of parenchymal lung diseases include adventitious breath sounds such as . The chest X-ray may provide radiographic evidence of ALL 20 Diagnosis and Source of Sepsis

In early literature, sepsis-induced lung disease was seen pathologically with increased interstitial edema and bronchopneumonia/lung inflammation. Physiologically, an increased shunt fraction likely related to increased parenchymal atelectasis, and decreased compliance primarily related to non- cardiogenic lung edema was observed. This lung disease initially was termed “shock lung” [29]. This was later termed acute respiratory distress syndrome (ARDS), or ALI, produced by lung inflammation related to the inflammatory-mediated injury as a result of a septic insult. This entity is defined by the presence of new bilateral infiltrates on the chest radiograph, ratio less than or equal to 300 in ALI or a ratio less than equal to 200 in ARDS, and no evidence of left atrial hypertension.

Renal Findings

Sepsis-induced renal dysfunction is evidenced by decreased urine output, elevated blood urea nitrogen (BUN) and creatinine, and uremia. In previous studies of bacteremic patients, approximately 24% developed acute renal failure, defined as a doubling of their creatinine. The mortality for bacteremic patients who developed acute renal failure was 50% [30]. Causes of sepsis- induced renal dysfunction include decreased glomerular filtration. Systemic hypotension and vasodilatation may contribute to some of this dysfunction by decreasing renal perfusion. However, afferent and efferent vasoconstriction also occurs as the renin-angiotensin system upregulates during sepsis- induced . Cortical ischemia (anuria suggests bilateral kidney involvement) secondary to sepsis-induced disseminated intravascular coagulation (DIC) thrombosis may also play a role [31]. A direct mediator effect on kidney function is also possible. Documentation of acute tubular necrosis as a cause of renal dysfunction is not supported by the literature. Some causes of sepsis such as endocarditis can produce independent renal injury. Other pyogenic foci have been shown less frequently to produce glomerular disease. However, this glomerulopathy does not correlate with the immune complex disease that has been demonstrated in endocarditis [32].

Gastrointestinal Findings

Ileus (intestinal obstruction secondary to neuromuscular bowel dysfunction) may be adynamic and occur in the critically ill, including sepsis. Although usually benign, this obstruction may delay feeding, decrease medication absorption or cause aspiration or perforation [33]. The causes are usually Sepsis 21

multifactorial, but medications and electrolyte abnormalities (e.g., hypokalemia) must be considered. Liver dysfunction during sepsis is, in some reports, as low as 0.6%, or as high as 50-60% [34]. Higher percentages of liver dysfunction are seen in post-surgical patients with peritonitis. Commonly, a cholestatic picture emerges with hyperbilirubinemia. While Gram-negative sepsis is frequently the cause, hyperbilirubinemia was initially described in patients with Streptoccocus pneumoniae infection [35]. The pediatric population is more prone to develop liver dysfunction with sepsis. Neonates are believed to have decreased rates of bile salt formation and a smaller pool of biliary salts. Speculation exists that Gram-negative sepsis may also cause decreased biliary flow through its interaction with membrane bound ATPase [36]. Pathologically, there appears to be no evidence of macroscopic parenchymal necrosis. However, microscopically, intrahepatic cholestasis, Kupffer cell hyperplasia, focal liver cell necrosis, portal tract inflammation, and venous congestion are found in many cases [37]. The laboratory abnormalities reflect the intrahepatic cholestasis observed under the microscope and are manifested primarily by rises in direct bilirubin. Commonly, there may be a minor increase in transaminases, but usually no more than two to three times normal, not approaching the rise seen in hepatitis or ischemic liver disease. Alkaline phosphatase typically rises, again not more than three times that of normal values. These altered values will return to normal, usually within two to six weeks with the resolution of sepsis [34]. Hyperbilirubinemia, predominantly direct in nature, is found on laboratory evaluation. The indirect fraction is usually low implying liver dysfunction as the cause of hyperbilirubinemia. Hemolysis however may occur with sepsis, as evidenced by an elevated indirect bilirubin. Clostridium perfringens infection for example has been linked to severe intravascular hemolysis [38].

Dermatologic Findings

Dermatologic manifestations of infection are plentiful. The causes of dermatologic manifestations include: 1) DIC and associated coagulopathy; 2) direct invasion of blood vessels by circulating bacteria; 3) immune complex formation and vasculitis; and 4) emboli from endocarditis [39]. Toxic shock syndrome also produces a characteristic blanching diffuse malar rash (Figure 3). The skin changes associated with DIC include acrocyanosis, defined as peripherally located grayish of the extremities, symmetrical peripheral gangrene (SPG), and purpura fulminans [39]. Commonly caused 22 Diagnosis and Source of Sepsis

by Gram-negative bacteria, especially Neisseria meningitidis, these skin changes include purpura, ecchymoses, or acrocyanosis (Figure 4). Progression to SPG may occur with ischemic necrosis of the peripheral extremities. Purpura fulminans is characterized by coagulopathic hemorrhage into necrotic skin lesions. All these changes are related to DIC with no histologic evidence of bacterial invasion. Sepsis 23

Direct vascular infiltration by circulating bacteria can also occur. This can result in similarly purpuric lesions that were initially macular. These purpuric lesions can be seen in meningococcemia as well as in overwhelming Pseudomonas bacteremia with invasion of the medial and adventitious walls of veins resulting in hemorrhage into the dermis. These hemorrhagic vesicular lesions soon rupture, leaving an ulcer with a necrotic base. Ecthyma gangrenosum, as these lesions are known, typically develop with bacterial invasion of the endothelium and minimal presence of inflammatory cells. The lack of inflammatory reaction can be explained by the fact that most of the victims are neutropenic. Gram-negative bacteria may be isolated from these lesions on biopsy. Candida infections can produce lesions similar to ecthyma gangrenosum [40]. There are also vasculitic skin lesions that develop from immune complex formation. These lesions are seen in more chronic forms of N. meningitidis as well as N. gonorrhea. Arthritis, nephritis, and myocarditis may be associated with these skin lesions. Lesions usually form seven days or more after infection, with antibiotic therapy already instituted. In disseminated gonococcemia, lesions develop in association with polyarthritis. Routinely, these lesions are maculopapular in nature, but can also be purpuric, pustular or ulcerated. On biopsy, microorganisms are rarely found in these lesions. Instead, antigens and antibodies can be isolated in the vessel walls. Immune complex formation may be found with endocarditis, manifesting with Osler nodes, Janeway lesions, purpura, and erythematous lesions. Finally, toxins such as those found with Staphylococcus and Streptococcus can generate well-characterized dermatologic lesions. Toxic shock syndrome exhibits a diffuse, macular erythroderma that blanches with pressure, subsequently fading, and desquamating in three to four days [41]. Histologically, these lesions reveal perivascular edema and lymphatic perivascular cuffing. Immune complex deposition is rarely present.

Metabolic Findings

Both hypoglycemia and hyperglycemia may be seen in sepsis. Low blood sugar occurs early and, as previously mentioned, may be a cause of mental status changes such as confusion, obtundation and seizures. While the possible mechanisms of this abnormality are diverse, a common theme in the literature involves both impaired production by the liver and increased uptake at the tissue level. Carbohydrate metabolism as influenced by counterregulatory hormones may also play a role. Recent studies in septic rats have shown increased uptake of glucose at several sites (liver, spleen, lung and ileum) in both euglycemic and hypoglycemic rats as well as 24 Diagnosis and Source of Sepsis

insulopenic rats, the latter suggesting that factors other than insulin affect glucose uptake [42]. In other studies, the suggestion has been made that impaired gluconeogenesis in the liver coupled with increased peripheral uptake accounts for the hypoglycemia observed in sepsis [43]. Hyperglycemia is seen commonly during sepsis. Because of the stress state associated with infection, increased epinephrine, corticosteroids, and glucagon drive the production of glucose, overwhelming the counterregulatory hormone insulin. Specifically, the elevated glucagon/insulin ratio drives hepatic gluconeogenesis and insulin resistance peripherally, leading to an overall overproduction of glucose. This excess glucose production counteracted by decreased peripheral utilization leads to hyperglycemia. Epinephrine both inhibits insulin release through the interruption of insulin exocytosis, an mediated action, while also increasing hepatic production through stimulation [44]. In addition, the peripheral lactate production funnels into the liver, producing glucose through the Cori cycle further contributing to the hyperglycemia observed. In addition, increased protein catabolism leads to the production of amino acids, which provide substrate in the liver for gluconeogenesis as well [453].

Hematologic Findings

The most common hematologic manifestation of sepsis is leukocytosis. However, the absence of abnormalities in white blood cell (WBC) count cannot be used to reliably exclude sepsis. The rise in WBC count is commonly over 10,000 Infection by bacterial source may further be identified by a left shift in the WBC count. Left shift refers to neutrophil predominance in the WBC count, or a rise in the immature form of neutrophils, called ‘bands’. Bandemia indicates a shift in marrow production towards neutrophil formation to combat infection. With the stress of infection, some of the WBCs may transit into the intravascular compartment before full maturation, thus increasing the number of bands in the peripheral circulation. On the other end of the spectrum, one can also see leukopenia as a presenting sign of sepsis. Leukopenia represents sepsis-mediated marrow suppression, leading to a paradoxical decrease in WBC count. An ominous sign, leukopenia is also associated with increased mortality during sepsis. Thrombocytopenia is another hematologic manifestation of sepsis. Thrombocytopenia occurs in 18%-50% of septic patients [46,47]. There are two possible causes for this manifestation: the evolution of platelet- associated immunoglobulins, or the development of DIC. Immune-mediated thrombocytopenia is associated with higher levels of platelets (> 50,000/ml) Sepsis 25

than are found in DIC (< 50,000/ml) [48]. DIC, on the other hand, indicates severe disease. Other laboratory signs of DIC include an elevated prothrombin time, an elevated D-dimer (indicating clot formation with increased fibrin split products), and a decreased fibrinogen (signaling consumption of clotting factors). Small quantities of thrombin production may be triggered by sepsis and produce thrombocytopenia without other manifestations of DIC.

SPECIAL CONSIDERATIONS IN THE GERIATRIC POPULATION

The geriatric population represents the most rapidly growing segment in the medical community. Of greater concern, sepsis has become the third leading cause of death in the elderly [49]. The elderly are at great infection risk due to immune system compromise, increased number of comorbid illnesses, and residence in chronic care facilities. Commonly, the geriatric population lacks the appropriate increase in WBC count making the diagnosis of sepsis in this population very difficult. Atypical presentation of infection is often observed. Fever may be absent or blunted in the elderly, and may infer poor prognosis. The inability of the elderly to mount a febrile response may be related to: 1) the presence of older macrophages unable to release endogenous pyrogens; 2) an anterior hypothalamus that cannot respond appropriately to pyrogens; or 3) an individual who cannot mount the proper physiologic response to endogenous pyrogens such as cytokines [50]. Therefore, the presence or absence of fever is unreliable in ruling in or out the diagnosis of sepsis in the elderly.

SPECIAL CONSIDERATIONS IN THE PEDIATRIC POPULATION

As with the elderly, identification of sepsis in the neonatal and pediatric populations can perplex the best of physicians. In this instance, relying on parental observation becomes critical in the history. Once again, vital signs play a critical role in diagnosing sepsis in the infant. From studies done in neonatal , the most important clinical signs coinciding with severe infection included respiratory status, mental affect, and decreased peripheral perfusion [51]. Good quantitation of urine output, and judicious use of arterial blood gases assist the clinician in assessing perfusion status. Decreased urine output, metabolic acidosis or respiratory alkalosis are all signs of hypoperfusion and may occur despite maintenance of normal blood 26 Diagnosis and Source of Sepsis

pressure. This is more likely to occur in the pediatric population than in adults.

DIFFERENTIAL DIAGNOSIS OF SEPSIS AND SEPTIC SHOCK

In treating the patient with sepsis, the physician must generate a differential diagnosis to include clinical entities that present similarly to sepsis [52]. This differential diagnosis includes illnesses that may also manifest some combination of tachypnea, tachycardia, hyperthermia, hypothermia, and leukocytosis. Non-infectious causes of fever include toxins such as cocaine or salicylates, thyroid storm, adrenal insufficiency, neuroleptic malignant syndrome, environmental heat injury, and hypothalamic injury. Systemic illnesses that can present similarly to sepsis include collagen vascular disease or vasculitic syndromes, solid and blood borne neoplasms, and drug overdose and toxins. Those disease syndromes associated with shock and acidosis include acute myocardial infarction, , acute hemorrhage, adrenal insufficiency, and anaphylaxis or drug reaction.

CONCLUSION

In conclusion, the best diagnostic tool for sepsis remains suspicion. Other than fever, many of the other commonplace signs of sepsis have a broad differential; therefore, the clinician’s intuition, history, and physical examination provide important clues to the diagnosis of sepsis. In addition, laboratory evaluation provides clues that may be hidden from the physical examination. Accurate early diagnosis and institution of empiric medical therapy is critical. Vital signs are important clues and must be combined with history, physical examination, and laboratory values to guide the clinician in the diagnosis and early treatment of sepsis.

REFERENCES

1. Harris RL, Musher DM, Bloom K, et al. Manifestations of sepsis. Arch Intern Med 1987; 147:1895-1906 2. Parker MM, Parillo JE. Septic shock: hemodynamics and pathogenesis. Arch Intern Med 1987; 147:1895-1906 3. Tandberg D, Sklar D. Effect of tachypnea on the estimation of body temperature by an oral thermometer. N Engl J Med 1983; 308:945-6 Sepsis 27

4. Dinarello CA, Cannon JG, Wolff SM, et al. Tumor necrosis factor is an endogenous pyrogen and induces the production of interleukin-1. J Exp Med 1986; 163:1433-1446 5. Dinarello CA, Wolff SM. Pathogenesis of fever in man. N Engl J Med 1978; 298:607- 612 6. Kluger MJ, Ringler DH, Anuer MR. Fever and survival. Science 1975; 188:166-168 7. Vaughn LK, Veale WL, Cooper KE. Antipyresis: Its effect on mortality rate of bacterially infected rabbits (abstract). Fed Proc 1980; 39:1077 8. LeGall J, Lemeshow S,Leleu G, et al. Customized probability models for early severe sepsis in adult intensive care patients. JAMA 1995; 273:644-650 9. Bryant RE, Hood AF, et al. Factors affecting mortality of gram negative rod bacteremia. Arch Intern Med 1971; 127:120-128 10. Clemmer TP, Fisher CJ, Bone RC, et al. Hypothermia in the sepsis syndrome and clinical outcome. Crit Care Med 1992; 20:1395-1401 11. Simmons DH, et al. Hyperventilation and respiratory alkalosis as signs of gram negative bacteremia. JAMA 1960; 174:2196-2199 12. Astiz, ME, Rackow EC. Septic Shock. Lancet 1998;351:1501-5 13. Kaehny WD. Respiratory acid-base disorders. Med Clin North Am 1983; 67:915-928 14. Parker MM, et al. Serial cardiovascular variables in survivors and nonsurvivors of human septic shock: heartrate as an early predictor of prognosis. Crit Care Med 1987; 15:923-9 15. Parrillo JE. Shock syndromes related to sepsis. In: RL Cecil, L. Goldman, JC Bennett (eds) Cecil Textbook of Medicine. WB Saunders, Philadelphia, 1970, pp. 496-501 16. Parrillo JE. Pathogenetic mechanisms of septic shock. N Engl J Med 1993; 328:1471-77 17. Abraham E, Shoemaker WC, Bland RD, et al. Sequential cardiorespiratory patterns in septic shock. Crit Care Med 1983; 11:799-803 18. Eidelman LE, Putterman D, Putterman C, et al. The spectrum of septic encephalopathy: definitions, etiologies, and mortalities. JAMA 1996; 275:470-473 19. Sprung CL, Peduzzi PN, Shatney CH, et al. Impact of encephalopathy on mortality in sepsis syndrome. Crit Care Med 1990; 18: 801-806 20. Freund HR, et al. Amino acid derangements in patients with sepsis: Treatment with branched chain amino acid rich infusions. Ann Surg 1978; 188:423 21. Papadopoulos MC, Davies DC, Moss RF. Pathophysiology of septic encephalopathy: a review. Crit Care Med 2000; 28:3019-3024 22. Vincent JL, Dufaye P, Barre J, et al. Serial lactate determinations during circulatory shock. Crit Care Med 1983; 11:449-451 23. Levy B, Sadoune LO, Gelot AM, et al. Evolution of lactate/pyruvate and arterial ketone body ratios in the early course of catecholamine-treated septic shock. Crit Care Med 2000; 28 114-119 24. Essen F, Lutfi T, Cakar N, et al. Comparison of gastric intramucosal pH measurements with oxygen supply, oxygen consumption, and arterial lactate in patients with severe sepsis. Adv Exp Med & Biology 1996; 388:521-531 25. Ince C, Sinaasappel M. Microcirculatory oxygenation and shunting in sepsis and shock. Crit Care Med 1999; 27:1369-1377 26. Vary TC, Siegel JH, et al. Effect of sepsis on activity of pyruvate dehydrogenase complex in skeletal muscle and liver. Am J Physiol 1986; 250:E634-640 27. James HJ, Luchette FA, McCarter FD, et al. Lactate is an unreliable indicator of tissue hypoxia in injury or sepsis. Lancet 1999; 354:505-508 28. Gore DC, Jahoor F, Hibbert JM, et al. Lactic acidosis during sepsis is related to increased pyruvate production, not deficits in tissue oxygen availability. Ann Surg 1996; 224:97- 102 29. Clowes GH. Pulmonary abnormalities in sepsis. Surg Clin North Am 1974; 54:993-1011 30. Rayner BL, Willcox PA, Pascoe MD. Acute renal failure in community-acquired bacteraemia. Nephron 1990; 54:32-35 28 Diagnosis and Source of Sepsis

31. Coratelli P, Passavanti G, Giannattasio M, et al. Acute renal failure after septic shock. Adv Exp Med Bio 1987; 212:233-243 32. Beaufils M, Morel-Maroger L, Sraer JD, et al. Acute renal failure of glomerular origin during visceral abscesses. N Engl J Med 1976; 295:185-9 33. Liolios A, Oropello JM, Benjamin E. Gastrointestinal complications in the intensive care unit. Clin Chest Med 1999; 20:329-345 34. Ledgerwood A. Hepatobiliary complications of sepsis. Heart Lung 1976; 5:621-623 35. Zimmerman HJ, Thomas LD. The liver in pneumococcal : Observations in 94 cases of liver function and jaundice in pneumonia. J Lab Clin Med 1950; 35:556-567 36. Zimmerman HJ, Fang M, et al. Jaundice due to bacterial infection. Gastroenterology 1979; 77:362-374 37. Banks JG, Foulis AK, Ledingham IM, et al. Liver function in septic shock. J Clin Pathol 1982; 35:1249-52 38. Batge B, Filejski W, Kurowski V. Clostridial sepsis with massive intravascular hemolysis: rapid diagnosis and successful treatment Intensive Care Med 1992; 18:488- 490 39. Kingston ME, Mackey D. Skin clues in the diagnosis of life-threatening infections. Rev Infect Dis 1986; 8:1-11 40. Bodey GP, Luna M. Skin lesions associated with disseminated candidiasis. JAMA 1997; 229: 1466-1468 41. Tofte RW, Williams DN. Clinical and Laboratory manifestations of toxic shock syndrome. Ann Intern Med 1982; 96:843-7 42. Lang CH, Dobrescu C. Sepsis-induced increases in glucose uptake by macrophage-rich tissues persist during hypoglycemia. Metabolism 1991; 40:585-593 43. Fikins JP, Figlewicz DP. Increased insulin responsiveness in endotoxicosis. Circ Shock 1979; 6:l-6 44. Weissman C. The metabolic response to stress: an overview and update. Anesthesiology 1990; 73:308-327 45. Barton R, Cerra FB. The hypermetabolism multiple organ failure syndrome. Chest 1989; 96:1153-1160 46. Poskitt TR, Poskitt PK. Thrombocytopenia of sepsis: the role of circulating IgG- containing immune complexes. Arch Intern Med 1985; 145:891-894 47. Kelton JG, Neame PB, Gauldie J, et al. Elevated platelet-associated IgG in the thrombocytopenia of septicemia. N Engl J Med 1979; 300:760-764 48. Neane PB, Kelton JG, Walker IR, et al. Thrombocytopenia in septicemia: the role of disseminated intravascular coagulation. Blood 1980; 56:88-92 49. Krieger BP. Sepsis and Multiorgan Failure: Sepsis in the Geriatric Age Group. Williams and Wilkins, Baltimore 1997 50. Cunha B. Infectious Diseases in the Elderly. PSG Publishing Company, Littleton, 1988 51. Bonadio WA, Hennes H, Smith D, et al. Reliability of observation variable in distinguishing infectious outcome of febrile young infants. Pediatr Infect Dis J 1993; 12:111-14 52. Dellinger RP. Current therapies for sepsis. Infect Dis Clin North Am 1999; 13:495-509