I am George Karam, an infectious disease physician with the LSU School of Medicine in New Orleans based in Baton Rouge. I'd like to look with you at some severe infections that are in the bacterial, fungal, and viral worlds with an emphasis on situations that might present themselves as a single best answer on an examination. Our first patient is a 62-year-old freshwater fisherman with cirrhosis due to alcohol. He presents to the emergency department with a three-day history of fever, diffuse abdominal pain, and pain over his left upper arm. The pain has progressively worsened since its onset and the patient develops diarrhea. On exam, he's in marked distress with a temperature of 39, a pulse of 132, and a blood pressure of 98 over 56. His abdomen is distended with tenderness in the right lower quadrant. His left shoulder is erythematous and indurated with crepitus that extends towards the antecubital fossa. The white count is 14,400 with 92% polys and his hematopoietic is 31% with an MCV of 69. Abdominal X-ray shows a pattern of obstruction with dilated small bowel loops and multiple air fluid levels. The most likely pathogen to cause this disease process is Streptococcus pyogenes, Eremonas hydrophila, Vibrio vulnificus, or Clostridium septicum. The four pathogens listed as the answers are all from the 2014 Infectious Disease Society of America guideline for the causes of necrotizing monomicrobial skin and skin tissue infection. If we look at the patient, there are two things that are really important. First, his link to water and it is fresh water, not salt water. And secondly, this presentation of a microcytic anemia with an abdominal X-ray that shows pattern of obstruction consistent with perhaps something like a carcinoma of the colon. If we think about colon cancer predisposing to infection with organisms like Streptococcus galiliticus, the old Strep bovis, or Clostridium septicum, and now we hear about this story of a man who seems to have a necrotizing skin and skin structure infection, pulling all this together, the most likely answer would be Clostridium septicum. In looking at the organisms that cause necrotizing monomicrobial infections, an important association is exposure to water. Eremonas is classically associated with not only fresh, but also brackish water. The injuries usually follow trauma in the water and most often occur on the extremities. Most classically is going to be cellulitis, but this can progress to myonecrosis with or without gas formation. And the guideline speaks about doxycycline and ciprofloxacin is the regimen of choice. In contrast, in exposure to salt water, Vibrio volnificus becomes the most likely pathogen. The classic group of patients who get this are those with chronic liver disease. That can include more than just alcohol-associated cirrhosis or steatohepatitis and can be iron overload states like hemochromatosis. The regimen is again going to be doxycycline-based, but this time with the third generation cephalosporin. Very importantly is the role for surgery to debride the necrotic tissue because without that extensive debridement, successful therapy does not very commonly occur. On the list is Clostridium septicum, but that must be considered in the context of Clostridium perfringens. There's been a classic teaching over years that patients don't die in the first 24 hours of surgery from an infection, but there are some classic breaches of that tenet. Clostridium perfringens is one, along with group A strep and toxic shock. Clostridium perfringens has the ability to cause gas gangrene, usually associated with trauma. But an important feature of Clostridium perfringens is the ability to produce an alpha toxin called lethocinase, and lethocinase destroys the membrane of both white blood cells and red blood cells. Because of this, if you aspirate fluid from the site of surgery and you see lots of grand positive rods but no white cells, perhaps that is because the white cells were destroyed by lethocinase from Clostridium perfringens. The other association has been with massive amylosis. And as our patient taught us, with GI malignancy, a risk for Clostridium septicum to get into the blood, seed structures like skin, and present as a necrotizing skin and skin structure infection. Of the pathogens that can cause skin and soft tissue infections, Staph aureus is a classic, and there's some important features about that. The 2014 guidelines of the IDSA divide the management of skin and skin structure infections into those that do not have purulence versus those who have purulence. This is the branch point from the guideline table. You'll notice in the non-purulence section, you bring up things like erysiplas, whereas with purulence, carbuncles and furuncles and abscesses were Staph aureus, and so the presence of pus should make one think about Staph aureus. Even though Staph aureus can cause tissue destruction, it's not characteristically associated with the degree of shock or organ failure that we see with group A strep infections. So purulence, but not as much organ dysfunction. An increased risk of Staph aureus in diabetics, probably because glucose control impairs the function of polystikil. The literature has said that if you look at patients with Staph aureus endocarditis, one in three will be diabetics, which leads to the very important property of endothelial adhesiveness. Staph aureus can stick to endothelial surfaces to cause endocarditis, and that's the reason for prolonged durations of therapy in patients who have diabetes and Staph aureus bacteremia. Whenever Staph gets into the blood, you can see things like the paraspinal space leading to epidural abscess and a real propensity to see muscle with psoas abscess being a characteristic presentation. We think about organisms destroying tissue or infecting tissue, but sometimes it's the production of a product by an organism that causes the disease. Toxic shock syndrome is such a representation. For years, we thought about it in the context of menstrual cycles, but the more contemporary view of toxic shock is in this category of non-menstrual toxic shock. Imagine a patient who has a surgical procedure with a little bit of erythema at the site. Staph aureus could proliferate, not destroy tissue or even give you inflammation, but produce the toxin. In a patient who has a lung infection, perhaps after influenza A and develops a Staph aureus pneumonia, large burden to bug to produce toxin. Mucosal surfaces, for example, the cervix with a cerclage, a little bit of erythema, Staph infection, the production of toxin. The bottom line is that when Staph has the ability to make toxin, it can present as toxic shock, as does group A streptococcus. When the term cellulitis comes up, very often there's an immediate association with infection, but cellulitis is really just erythema of the skin, and so there are some non-infectious things of skin that can result in skin redness. For example, muscle with an anaerobic infection will produce erythema of the overlying skin. Unusual bacteria like Bacillus anthracis, whenever they cause cutaneous infection may not give us the classic Staph group A strep presentation. But very importantly are the forms of fasciitis, and those can be divided into two. Those with mixed aerobes and anaerobes, the Forniase variety, and then the monomicrobial aerobic organisms like group A streptococcus. The classic teaching has been that Forniase produces pain out of proportion to the skin lesions that you see. If one were going to conceptualize Forniase gangrene, it's going to occur in patients with some degree of vascular insufficiency, often a diabetic, but maybe somebody with just peripheral vascular disease. And the location's going to be around the anus, because that's where anaerobes and aerobes have their access to the skin. So imagine then a perineal location in somebody with vascular insufficiency. Surgical debridement is key, along with antibiotics that cover a broad spectrum of organisms, including MRSA, Pseudomonas aeruginosa, and anaerobes. Our second patient is a 35-year-old man who lives in Nantucket, but recently traveled to Africa. And he goes to his emergency department five days after he returns home with complaints of fever and myalgias. His past history is significant for surgery following trauma sustained in a motor vehicle crash 12 years earlier. He does not appear to be acutely ill, and he's sent home with acetaminophen for his symptoms. But 12 hours later, he returns acutely ill with fever, hypotension, an altered level of consciousness, the skin lesion below, which is purpuriforminans, and the blood smear. The most likely pathogen to cause this clinical presentation is Plasmodium falciparum, Babesia microti, Neisseria meningitidis, and Streptococcus pneumoniae. The key feature of this patient's illness was a flu-like state, followed about 12 hours later by an acute life-threatening illness. His epidemiology of being from Nantucket raises the issue of Babesia. It's similar to Africa, raising the issue of malaria. But if one looks at the peripheral smear, that solid basophilic inclusion isn't consistent with the Maltese cross-form in Babesia or the Cygnet ring in malaria. That is a howl, jolly body indicative of the asplenic state. And the most classic organism to cause life-threatening infection in asplenic patients, in fact occurring in over 80% of the patients, is Streptococcus pneumoniae. So although any of these four pathogens would be a possibility, in this patient the most likely diagnosis is Streptococcus pneumoniae. We recently conceptualized Fournier's gangrene. Let's think about a conceptualization of Streptococcus pneumoniae. And I would propose that you think about it as a highly inflammatory organism that can really lead to profound illness on the basis of that inflammation. And there are four groups of people who are most at risk for that. The asplenic, as we just saw. And multiple myeloma, the lack of ability to form normal amounts of immunoglobulin for opsonization of the pneumococcus. And alcoholism, that poisoning of the defense mechanisms so pneumococcus in the airway can be left unfettered and get into the blood. And then with HIV infection, the loss of regulation of B-cells because of destruction of T-cells and the ability for a normal humoral immune response. Since the 1998 guidelines of the IDSA, it's been said that there are two organisms likely to kill you acutely from CAP, pneumococcus and Legionella. And the guideline goes on to say that in patients who don't respond well to bacteremic pneumococcal pneumonia, even when you're doing everything perfectly, one must think about streptococcus pneumonia because this inflammatory nature of the organism. So the presentation of an inflammatory state in these disease entities. I want you to think for a minute about a 59 year old man with COPD who's been on prednisone 5 milligrams a day for the last year, who now acutely develops fever, headache, vomiting and photophobia. You do an LP on this person and he's got 258 white blood cells with 91% polys. CSF grows Enterobacter cloacae. What are you thinking about the basis for this? We understand that Enterobacter is not a classic organism to cause meningitis. So the question that becomes important is might there be something else that predisposes to Enterobacter? We know that it can be a part of the enteric floor of the GI tract. So if something breached the integrity of the mucosa of the GI tract, did Enterobacter now have access into the bloodstream? In patients who've been on steroids, depression of cell mediated immunity may lead to infection with stronger loities. And so this is the presentation of strong loities infection leading to gram negative bacteremia and the gram negative bacteremia leading to meningitis. Which opens up our need to think about the approach to meningitis in acutely ill people. This is the concept map that I've constructed on the approach to CSF pleocytosis and I use that term because meningitis is nothing more than white cells in the CSF. Which implies that it can be both on a non-infectious as well as an infectious basis. The cell type becomes important, polys versus lymphs versus eosinophils. I would suggest that you forget about the eosinophils other than thinking about an entity like maybe angiostrongylus. In patients who go to the Caribbean, eat Caesar salad, the basis for a New England Journal report. And as we think about lymphocytes, the core thing is, is the glucose normal or not normal? For years people have said low glucose is related to bacterial consumption. That is incorrect. Low glucose is related to an impaired transport across the meninges. And that impaired transport may be due to inflammation by bacteria. It may be due to infiltration of granulomas and things like TB or fungal meningitis. Or it may be the infiltration of the meninges due to cancer cells. But it's impaired transport. And as we think about the differential, notice that under the normal glucose category, infectious versus non-infectious causes. And in the low glucose, again, infectious versus non-infectious. So as you think about your approach for the exam to meningitis, I would propose that you think about this map, which will then take you to the important considerations. The therapy of meningitis is very good fodder for an exam. And I've divided it into these three categories based on the way the literature reports it. Within the normal host, the two most important organisms to consider are Streptococcus pneumoniae and Neisseria meningitidis. But once patients develop some defect in their cell-mediated immunity, they become unable to handle intracellular organisms and Listeria monocytogenes is an intracellular organism. Keep in mind that age over 50 is associated with defective cell-mediated immunity. And the third trimester of pregnancy is associated with a transient depression in cell-mediated immunity. So subtle clues beyond things like steroids and transplant medicines are going to be things like age over 50 and third trimester of pregnancy. And then finally, in individuals who've had penetrating head trauma, undergone neurosurgery or have a CSF shunt, we bring in MRSA and Pseudomonas aeruginosa. Those considerations become the basis for treatment. These are from recent guidelines that look at the therapeutic regimens. So in people where it's pneumococcus, meningococcus, mycomicin plus a third generation cephalosporin. Age over 50 because of the depressed cell-mediated immunity or any other classic reason for depressed cell-mediated immunity, the addition of ampicillin because cephalosporins do not cover Listeria. And then finally, where we bring in MRSA or Pseudomonas, the addition of vancomycin and either cefepime or ceftazidime or miripenem. But notice that even though some of those agents are recommended for the treatment, the meningitis due to Pseudomonas, they have not been FDA approved for meningitis, but they remain as our guideline options. Very easy to ask on an exam is a question about people who have recurrent neisserial infections including neisserial meningitis. When people have looked at that, the question is really testing the understanding that deficiencies in the late components of complement, C5 to C9, have the ability to predispose. We also know that certain monoclonal antibody regimens that target complement are now the predispositions to patients who may have now gram-negative meningitis. And then finally, I mentioned to you in the patient, the ability of strongyloides to breach the mucosa of the GI tract, let enteric gram-negative rise get into the bloodstream, that can then lead to gram-negative meningitis. So two situations, but keep in mind monoclonal antibody therapies that target complement is a predisposition to gram-negative bacteremia. Among life-threatening infections of the nervous system, there are two that I consider as being the triad diseases, which means that they classically present with three things. With spinal epidural abscess, it is fever, point tenderness over the spine, and focal neurologic deficits. But there is not a great reliability of defining where the lesion is based on just where the tenderness is, and so imaging of the entire spine becomes important. It's not understood about whether the abscess compresses venous outflow and then results in edema in the cord or compresses arterial supply into the cord, resulting in ischemia or just literally pushes on the cord to produce necrosis. But the bottom line is that it may be an acute process leading to irreversible damage within a matter of hours. So urgency is of paramount importance. MRI is going to be the diagnostic study of choice. Herpes simplex encephalitis is the triad of an acute illness with fever and focal neurologic deficits. But you're likely to be asked about focality in the form of some sensory abnormality. So for example, funny smell, being temporal lobe, abnormal vision, indicative occipital lobe. PCR on the CSF is highly important in the diagnosis, and the only A1 recommendation for viral therapy for a viral infection is going to be acyclovir for HSV. Our third patient is a 41-year-old homeless man with a seven-month history of sepsis associated with a forearm subcutaneous abscess that grew iconella corrodens. And he now gets admitted with a history of blurred vision and difficulty swallowing that began three days before admission. Over the past two days, the patient has been complaining of difficulty holding his head upright and moving his shoulders. Over the past eight hours, he's been increasingly having trouble taking deep breaths. On presentation to the emergency department, he has impending respiratory failure and requires intubation. His physical exam is notable for four centimeters subcutaneous abscesses, each measuring less than three centimeters on his left forearm. Based on this presentation, the most likely diagnosis in this patient is fasciitis due to streptococcus pyogenes, bacteremia with iconella corrodens, neuritis caused by clostridium botulinum, or embolic disease resulted from saplococcus aureus endocarditis. In this patient, there is a predisposition to each one of these disease processes. But as we think about this guy's story, what we find is that he's got a descending paralysis syndrome. And the descending paralysis syndrome is associated probably with his injection drug use where he injected black heroin, the spores of clostridium, and this is going to be the neuritis of clostridium botulinum. In the approach to infectious diseases, I would challenge you to think about one of three processes. Is it the organism itself either causing destruction as occurs with pneumococcus or an inflammatory response? We heard about it recently in pneumococcus, I'm sorry, tissue destruction in group A strep. The inflammatory response due to pneumococcus, but it can also occur with pneumocystis here in Vecchi. Is it toxin production? We saw the toxic shock syndrome. Keep in mind that our classic disease now is going to be clostridium difficile colitis, but it could be the shiga toxins. Or is it going to be an immunologic reaction, the post-infectious glomerulonephritides or rheumatic fever or very important, something like leptospirosis. And botulism is going to be an example of a toxin mediated problem. The epidemiology of the neuritis due to clostridium species can be best approached by looking at what the species of clostridium it is. Is it tetani or botulinum? Tetanus producing characteristically a spasted paralysis, whereas botulinum is a descending flaccid paralysis. The most recent epidemiology for both of these has been injection drug users who shoot subcutaneously. Their product is often black heroin that's contaminated with the spores of clostridial species and they can both present with a neuritis syndrome. The approach to the neuritis is whether it's a direct effect on the nerve, which is what happens with diphtheria. But when we think about both botulism and tetanus, it's an indirect effect. And for testing purposes, is there a role for antitoxin? There is for diphtheria and there is for the adult variant of botulism, not the childhood or the infant form. On the other hand, for tetanus, no role for antitoxin, but a role for toxoid. West Nile encephalitis is important because of its property of being neurotrophic, which means that it has the ability to affect neurologic tissue and cause disease. We would love to think that it's always going to present as meningitis, but about half of the time it may present as only a stroke-like syndrome. And so in older individuals with no classic risk for stroke, the possibility of West Nile presenting as a stroke syndrome. Elderly patients much more likely to irreversible neurologic damage. Since it was described as a disease state, West Nile encephalitis has been compared to the acute flaccid, in its acute flaccid paralysis form, has been compared to Guillain-Barre syndrome. This table gives you the discriminators. Guillain-Barre as a demyelinating disease is usually symmetric and it is usually associated with pain because of the loss of myelin. In contrast, West Nile acute flaccid paralysis is usually non-symmetric and because it's not demyelinating, does not have pain. On the other hand, West Nile acute flaccid paralysis has a CSF pleocytosis, Guillain-Barre does not. One of the classic pathogens in the nervous system is going to be rabies that begins with a bite, spreads up peripheral nerves, gets into the cord, goes into the brain, and hence an encephalomyelopathy. The myelopathy presenting as pain followed by the encephalopathy is cognitive dysfunction. And so rabies usually will present as pain in an extremity before the nervous system. One of the characteristic features of rabies is the autonomic instability. And a feature that should make one think about autonomic instability is hypersalivation. They can also develop hypertonicity. And then very importantly, once it moves from the brain it goes into the salivary glands which says there's a risk of spread to individuals and so a role for post-exposure prophylaxis. Keep in mind the last bullet, the CDC statement that says it's important to consider rabies in any patient with an unexplained acute encephalitis. Spirochetes have the ability to infect the nervous system in the form of syphilis, but that's usually stroke in a young person or as Lyme disease which is characteristically a facial nerve palsy that can be bilateral. But when it's going to be life threatening it usually is going to be as leptospirosis. The initial phase is the spirochete and then people get better and then they get really sick due to the immunologic phase. So an example of that immunologic injury and the triad of organ involvement is going to be brain, liver, and kidney. With rickettsia, Rocky Mountain spotted fever following a tick bite in somebody who's from the east coast of the United States, or epidemic typhus which may be associated with flying squirrels, but the one most likely to give you really severe disease is going to be ehrlichiosis. Ehrlichia is going to present with the triad of low white cells because of the intracellular location of the organism, thrombocytopenia, and elevated liver enzymes. Ration less than 20% of people but an important consideration in severe nervous system infection. In the gastrointestinal tract there are several mechanisms by which life-threatening diarrhea develops. In terms of invasion, cholera is the prototype and the guideline talks about dehydration, not just intravascular volume depletion, but dehydration is the leading cause of death. So fluid resuscitation of paramount importance. There's an ability to injure intestinal mucosa in people who are neutropenic. So you can imagine as they get chemotherapy that destroys the mucosa, the flora from the gut involves now this non-protected part of the GI tract and they get neutropenic enterocolitis, sometimes called typhlitis. Clostridium difficile colonizes about 8% of normal people but up to 50% of people in long-term care facilities and they may not have any problems because C. difficile alone doesn't create diseases, the production of toxin. And we are now beyond E. coli 0157 H7 into a whole group of the Shiga toxins which have the ability to do the HUSTTP that we initially thought with E. coli. And if we're thinking of a classic immunologic complication, it's going to be Guillain-Barre with this demyelination following campylobacter infection. The urinary tract infections are generally not so life-threatening but there are three to consider that may present acute illness. Renal papillary necrosis is going to be sloughing of renal papillae usually with secondary bacterial infection following that. You can imagine that in a diabetic who becomes ischemic and sloughs their papilla. The classic teaching in medical school was that gas is formed by anaerobes but emphysematous pyelonephritis is an example of gas being formed by aerobic gram-negative bacteria and so emphysema in the kidney as a result of aerobic infection. The curve to your left is what happens with pyelonephritis and that is temperature goes down gradually over three days but there continue to be spikes in the first couple of days. In contrast to pyelonephritic abscess where there is a continuous fever because of inability for source control until the pyelonephritic abscess is identified. So the curves allow for the distinction between pyelonephritis versus perinephric abscess. In the intravascular category I like to think about where the infection is located keeping in mind that the clue to intravascular infection is persistent or ongoing bacteremia because the infection is sitting within the blood vessel is always seeding it and so is the infection on endothelial surfaces the classic example being endocarditis. Is it in the arterial wall which could be something like an aortic aneurysm within the veins and we're going to look at Lemire syndrome or is it transfusion associated. This is my list of infections that I think are classic for being life-threatening show up in the ICU and they're a single best answer on an exam. Yersinia enterocolitica has the ability to survive at 27 degrees centigrade an important number because that's the temperature of refrigerators and so when patients donate blood and they're infected with Yersinia that organism can live and so when they receive a blood transfusion are they infused with the bug or is it just the toxin of the bug both are possible but transfusion related infection due to Yersinia. In the arterial wall mycotic aneurysms have conglomerations of blood kind of like a blood agar plate and we know that the iron in the blood may be the source for certain organisms that love iron. Salmonella is a classic organism that loves iron and so a finding like Salmonella cholerae suison blood should make one think about an infected aortic aneurysm. Seperative thrombophlebitis is classically thought about people who have catheters but I'd like to think about it with you and people who don't have a catheter that might be pelvic thrombophlebitis, cavernous sinus, or portal vein and then finally very importantly the entity of jugular vein thrombophlebitis the Lemire syndrome or post-anginal sepsis. It's best represented by this picture. Exudated pharyngitis due to the virus where you breach the mucosa so mouth anaerobes have the ability to move retrograde and one of those mouth anaerobes is Fusobacteria necrophora and it moves into the jugular vein and you can see on the imaging obstruction of the jugular vein so now you have essentially an abscess sitting in the jugular vein. It obstructs flow from the head so they get you know proptosis and facial swelling they get tinnitus over the neck they embolize and they get organisms into their lung much like what we think about with Staph aureus bacteremia on a tricuspid valve and there's an associated entity of arthritis. This is really going to be managed best with antibiotic therapy no one is sure if you have to give anticoagulation for all of these patients. In the management of bloodstream infections the general rule is you take the catheter out but with one important exception the first bullet. It's been shown that you can treat coagulation negative Staphylococcal bacteremia so an organism like Staphylococcus epidermidis by leaving the catheter in place and giving antibiotics. On the other hand there's one exception in coagulation negative staph and that's the Staphylococcus lupinensis which is coagulation negative which should be managed exactly like Staphylococcus aureus. The world of head and neck infections I've divided into these two categories. Epiglottitis is no longer really a pediatric disease because haemophilus influenzae vaccine has eliminated many of those infections in children so we now see it as an infection in adults. It presents mainly in the super epiglottic area presentation of kind of drooling the ability to lean forward. H flu is surely a pathogen but we worry about things like Staphylococcus aureus about pneumococcus about meningococcus because the flora is what you can have in the posterior pharynx. In approaching fascial plane infections one may want to think about three spaces the submandibular space the parapharyngeal space and then the retropharyngeal space. The classic example in the submandibular space is going to be Ludwig's angina which is characteristically bilateral in nature and can progress to abscess formation that may require a surgical drainage. When one begins to think about what can involve the parapharyngeal space keep in mind paratonsillar cellulitis or when it advances to abscess the quincy abscess and what this has the ability to do is to rupture from the paratonsillar space into the parapharyngeal space and that's probably the important parapharyngeal infection that we're going to see in the clinical setting in adults. And then finally the lateral pharyngeal space is going to include in the anterior compartment mainly muscle but as you move posteriorly it's going to pick up things like blood vessels and nerves. The hallmarks of these is to cover the flora of the upper airway with the appropriate assessment for abscess and whether there's a need for surgical debridement. And the approach to fungal infections I would challenge you to think about yeast and moles remembering that kind of in the intermediate zone or the dimorphic fungi like histo, blasto, and cocci. The next patient is a 46 year old man with ulcerative colitis who's been treated for the last five years with systemic corticosteroids and now develops fever, abdominal pain, and guarding with rebound. On exploratory laparotomy he's got a perforated duodenal ulcer with evidence of peritonitis. Postoperatively the patient receives broad-spectrum antibiotics through a central venous catheter but his course is complicated by a persistent ileus. On hospital day five he begins complaining of visual floaters. One of two blood cultures drawn on postoperative day two and when the patient had a temperature of 37 now now grows Canada albicans. Noteworthy on physical exam or hemodynamic stability in a normal cardiac exam. The retinal exam by an ophthalmologist reveals findings consistent with chorioretinitis without either vitreitis or macular involvement. What's the most appropriate management for this patient? Intravenous fluconazole, an intravenous echinocandon, liposomal amphotericin B, or intravitreal amphotericin B. The guidelines for candida have talked about the role of echinocandins but one place that echinocandins do not go very well is into the retina. The azoles on the other hand do very well in the retina. This patient has the entity of candida endophthalmitis and what you see at about the three o'clock position is that retinal abscess where candida seated the retina grows into the vitreous wiggles and gives trouble and so agents that get into the retina would become important. In the treatment recommendations for candidiasis, beyond an echinocandin in people with retinal disease are going to be the azoles and so the approach to chorioretinitis is without vitreitis or with vitreitis but in both of those you can see a role for an azole. So although the management for candidemia usually is going to be with an echinocandin, the guidelines states do an ophthalmologic exam and if you see evidence of retinal involvement consider perhaps the lack of efficacy of a kind of candida moving forward. The world of the critical care doctor is one very often dealing with sepsis and we think about it as bacterial sepsis. A recent editorial talked about the role of viral sepsis and said that's really what COVID-19 is is viral sepsis and now we're going to think about candida sepsis. The two trials on this paper make one important common fact that if you're going to treat candida and prevent fungal sepsis you have to source control the problem and the source control is going to be usually the removal of a catheter. Interestingly there are some reports that say that in neutropenic patients the source of the fungemia may be from the GI tract not the catheter and sometimes leaving the option for remaining the catheter if they are neutropenic. Again that's probably done best with infectious disease consultation but the concept is catheter removement to decrease mortality and septic shock caused by candida. The world of aspergillus is best approached by whether the patient is neutropenic or not because neutropenic patients have a risk of aspergillus invading the blood vessel. They can then occlude the blood vessel completely and mimic pulmonary embolus. Hence this picture that looks like a pulmonary embolus with infarction but was really a neutropenic patient with aspergillus that obstructed blood flow and mimicked embolus. In fact some leukemia services say the development of a moptosis in someone with prolonged neutropenia should raise the issue of aspergillus infection. On the other hand in patients who are not neutropenic but have a defect in their cell mediated immunity. So for example somebody on steroids for graft versus host disease they're going to have an entity that looks exactly like regular bacterial pneumonia because there is no angioinvasiveness rather there's just going to be this pyogranulomatous inflammation and that can lead to an inflammatory necrosis. So two different presentations based on disease state. This patient 63 years old and had intravascular volume depletion secondary viral gastroenteritis. Got admitted to the hospital and then had a temperature to 38.5. Her IV site was erythematous so they removed the catheter and they cultured it and they drew two blood cultures and they didn't give any antibiotics and her fever went away in two hours. Two days later her GI problem is resolved. She's ready for discharge but the lab tells you that one of two blood cultures and the catheter tip are now both growing a methicillin sensitive staph aureus. What's the right thing to do? Do you repeat the blood cultures but withhold the antibiotics unless they're positive? Give her oral cephalexin for ten days, give her intravenous cephaloxylin for seven days or intravenous nafcillin for two days. The classic tenet in medicine is treat the patient not the number. I will tell you with staph aureus bacteremia this is a breach of that tenet and you always treat the number not the patient because of that property of endothelial adhesiveness and the risk of endocarditis and epidural abscess. The conceptual approach to staph aureus bacteremia probably is best considered about whether it is a complicated or an uncomplicated infection. To call an infection with staph aureus the blood uncomplicated it needs to be catheter related where when you remove the catheter both the fever and the bacteremia go away within 72 hours and that 72 hour number is really important. There should not be any prosthetic device in place. The patient shouldn't be immunocompromised. The guideline from 2009 said they should not be diabetic and there should be no evidence of endocarditis on trans esophageal echo and for those people when everything is perfect like that they can get just two weeks of parenteral antibiotics. On the other hand whenever we think about complicated disease it implies one of two things either that you've seeded some distal site or it's been going on for more than 72 hours. People question whether diabetes counts as classified it's complicated but the 2009 guideline stated it is and there's nothing new that says that it's definitely not and then people would consider immunosuppressive therapy or intravascular devices as me making it complicated. Notice those individuals get at least four weeks of therapy unless there's endocarditis or osteo and then it's six to eight weeks. Which brings us to the entity of duration of antibiotic therapy. For years we used to give antibiotic therapy for prolonged durations two weeks three weeks four weeks then we recently begun to learn that probably we can get by for much shorter periods of time. The paper that I'm citing for you is from 2011 but I've updated it with the more contemporary things and so what you see is that when you get to classic diseases like community-acquired pneumonia, hospital-acquired pneumonia, intra-abdominal infection and pyelonephritis, shorter durations of therapy really can be the rule. So we're usually talking more in the range of about seven days. Situations where durations of therapy may be better defined. We've already looked at the numbers for Staph aureus bacteremia with two weeks for uncomplicated four weeks or more for complicated. The guidelines say that for patients with candidemia without metastatic complications two weeks after documented clearance but they leave open the option for some of that being with oral therapy. Enterococcus like candida and like Staph aureus has endothelial adhesiveness so it has a propensity to involve heart valves but if you have enterococcal bacteremia without endocarditis it has been said that maybe therapy can be only five to seven days if the repeated blood cultures are negative at 24 hours. Once endocarditis occurs with enterococcus it's a four to six week course of therapy and recently two reports saying that seven days may be as good as 14 days with some of that is oral therapy when we think about gram-negative bacteremia. Although we worry about the gram negatives they don't have the property of endothelial adhesiveness and therefore they are less likely to cause endocarditis. And finally not everything that presents with sepsis is going to be infection because the systemic inflammatory response may occur on a non-infectious basis. This table from an emergency department talks about some things that can mimic infection and present as sepsis. What is missing on that non-infectious side is ischemic bowel disease which I think is an extremely important entity for an acute presentation that may be life-threatening. But we understand that when you make people intravascular volume deplete they try to keep blood close to its core so they vasoconstrict peripherally. When they vasoconstrict peripherally temperature may go up and they may actually have hyperthermia not fever. Entities like adrenal insufficiency have the ability to produce things like interleukin-6 which have the ability to give you fever. Give people either dead heart muscle or dead lung and they're going to have tissue now that's going to release interleukins and that can cause fever. The bottom line is not everything that presents with sepsis will be infectious and these are important mimickers if we add ischemic bowel to the list. And then finally infections with negative cultures. C. difficile is a toxin-mediated disease. Cryptococcus may not grow as rapidly as certain bacteria although it can definitely grow in a blood culture and we know with tuberculosis and viral encephalitis the likelihood of not growing. And so I hope that what this discussion of bacterial fungal and viral infections did was to give you an idea of some entities that may show up on the exam where there may be a single best answer. Thank you very much.

severe infections

pathogens

necrotizing infections

Plasmodium falciparum

meningitis

Candida albicans fungemia

intravenous fluconazole

source control