WORKUP	Section 5 of 10
Author Information Introduction Clinical Differentials Workup Treatment Medication Follow-up Miscellaneous Bibliography

Lab Studies:

• Blood cultures

• Obtain blood cultures for all patients at admission to demonstrate the causative organism responsible for infection. Negative blood culture results are also necessary to rule in pseudosepsis.

• Blood culture isolates might suggest the underlying disease process. Bacteroides fragilis suggests a colonic or pelvic source whereas Klebsiella species or enterococci suggest a gallbladder or urinary tract source more frequently than an intra-abdominal source.

• A CBC count usually is not helpful because of the numerous conditions that mimic sepsis (eg, pseudosepsis) that present with a leukocytosis of variable degrees of a left shift. Leukocytosis with a left shift is a nonspecific diagnostic finding. It is as common in noninfection as with infection.

• Other

• Obtain a urine Gram stain, urinalysis, and urine culture if urosepsis is suspected.

• If central IV line sepsis is suspected, then remove the line and send the tip for semiquantitative bacterial culture. If the catheter tip culture results are positive and demonstrate 15 or more colonies, and if the catheter tip isolate matches the blood culture isolate, an infection associated with the central IV line is diagnosed.

• Obtain a buffy coat of the white cells from peripheral blood stained by acridine orange, or use the Gram method to demonstrate bacteria responsible for the bacteremia or septic process. While the yield is low, stained buffy coat smears, if positive, are the best rapid test available to demonstrate organisms causing bacteremia. If the stained buffy coat smear is positive, it demonstrates the morphology of the bacteria causing the bacteremia, which provides rapid information on which to base empiric antimicrobial therapy.

Imaging Studies:

• Chest radiograph is important to rule out pneumonia and diagnose the following other causes of pulmonary infiltrates:

• Pulmonary drug reactions

• Pulmonary emboli

• Pulmonary hemorrhage

• Primary/metastatic pulmonary neoplasms

• Lymphangitic spread of malignancies

• Large pleural effusions

• Pneumothorax

• Hydrothorax

• Fluid overload

• Congestive heart failure

• Acute myocardial infarction

• Acute respiratory distress syndrome should suggest an intra-abdominal source (eg, acute pancreatitis).

• Ultrasound

• Obtain an ultrasound of the abdomen if biliary tract obstruction is suspected based on clinical presentation.

• The ultrasound in cholecystitis may show a thickened gallbladder wall or biliary calculi but no dilatation of the common bile duct.

• Stones in the biliary tract can be seen in patients with cholangitis, but the common bile duct will be dilated.

• An abdominal ultrasound is suboptimal for the detection of abscesses or perforated hollow organs.

• CT scan or MRI

• Use CT scan or MRI of the abdomen if a nonbiliary intra-abdominal source of infection is suspected through history or physical examination.

• CT scan and MRI are superior to ultrasound in demonstrating all lesions except those related to the biliary tract.

• An abdominal CT or MRI scan also is helpful in delineating intrarenal and extrarenal pathology.

• Gallium or indium scan: This has no place in the initial workup of these patients, since patients who are septic are acutely ill by definition, and rapid diagnostic tests (eg, CT scan or MRI of the abdomen, ultrasound of the right upper quadrant) are time-critical, life-saving interventions.

Other Tests:

• ECG: Obtain an ECG and cardiac enzymes in patients where the diagnosis of acute myocardial infarction (MI) is likely. Remember that certain patients may present with a silent, asymptomatic MI, which should be included in the differential diagnosis of otherwise unexplained fever, leukocytosis, and hypotension. Silent MIs are common in the elderly and in those who have had recent abdominal or pelvic surgery. They also occur frequently in individuals with alcoholism, diabetes, and uremic conditions.

Procedures:

• If a substantial pleural effusion exists, perform a thoracentesis for diagnostic purposes.

• Perform a paracentesis for patients with gross ascites.

• Swan-Ganz catheter

• Use data obtained via Swan-Ganz catheter to manage the fluid status of the patient and to assess left ventricular dysfunction in those patients with acute myocardial infarction.

• Do not use Swan-Ganz hemodynamic parameters to diagnose sepsis. While most patients with sepsis demonstrate an increased cardiac output with a low peripheral vascular resistance, the converse is not true. Most patients presenting with Swan-Ganz readings compatible with sepsis without a definite intravenous abdominal or genitourinary (GU) source do not have sepsis, but rather a pseudosepsis, as described above.

	MEDICATION	Section 7 of 10
Author Information Introduction Clinical Differentials Workup Treatment Medication Follow-up Miscellaneous Bibliography

Appropriate antimicrobial therapy depends on adequate coverage of the resident flora of the organ system presumed to be the source of the septic process. Empiric monotherapy regimens include imipenem, meropenem, cefoperazone, piperacillin/tazobactam, or sulbactam/ampicillin. Combination therapeutic agents include clindamycin or metronidazole plus either levofloxacin, aztreonam, trimethoprim/sulfamethoxazole (TMP-SMZ), or an aminoglycoside.

No one drug/regimen is superior to another. Alternative agents may be used alone or in combination providing they have a low reactive potential and a good adverse effect profile.

Drug Category: Antibiotics -- Used to treat various types of infection.

Empiric therapy for IV line infections

Since IV-line infections due to methicillin-resistant Staphylococcus aureus (MRSA) are infrequent, do not use vancomycin empirically for IV-line infections.

If coagulase-negative staphylococci are recovered from the blood, do not use vancomycin empirically to treat these patients, since this is a low-virulence organism. Treatment of coagulase-negative, staphylococcal central-line infection requires removal of the line and not empiric vancomycin therapy.

If a central-line due to MRSA or coagulase-negative staphylococcal infection cannot be removed for clinical reasons, then empiric suppressive vancomycin therapy is acceptable.

Minimize use of vancomycin in order to prevent the emergence of Enterococcus faecium, a vancomycin-resistant enterococcus (VRE).

Preferred monotherapy for IV line infections is imipenem, meropenem, cefoperazone, or cefepime.

Combination therapy should also be with an antistaphylococcal penicillin, eg, nafcillin or linezolid plus aztreonam, an aminoglycoside or a quinolone.

Empiric therapy for biliary tract infections (cholecystitis/cholangitis)

The main biliary tract pathogens are Escherichia coli, Klebsiella, or Enterococcus faecalis. Staphylococcal and anaerobic coverage are not needed in the biliary tract. Anaerobes are important only in diabetics with emphysematous cholecystitis due to Clostridium perfringens. Preferred monotherapy for biliary tract infections is with imipenem, meropenem, piperacillin, or cefoperazone.

Empiric therapy for intra-abdominal and pelvic infections

The main pathogens in the lower abdomen/pelvis are aerobic coliform gram-bacilli and B fragilis. Enterococci are permissive/opportunistic pathogens and do not require special coverage. Potent anti-B fragilis and aerobic gram-negative bacillary coverage are essential, in addition to surgical intervention when drainage or repair of intra-abdominal viscera is required.

Preferred monotherapy for intra-abdominal and pelvic infections is imipenem, meropenem, piperacillin/tazobactam, or ampicillin/sulbactam.

Preferred combination therapy for intra-abdominal and pelvic infections is clindamycin or metronidazole plus either aztreonam, levofloxacin, or an aminoglycoside.

Empiric therapy for urosepsis

The primary uropathogens are gram-negative aerobic bacilli, eg, coliforms or enterococci (E faecalis, not E faecium vancomycin-resistant enterococci [VRE]).

Pseudomonas aeruginosa, Enterobacter, and Serratia are rare uropathogens and are associated with urological instrumentation.

Preferred monotherapy for urosepsis due to aerobic gram-negative bacilli is with aztreonam, levofloxacin, third or fourth generation cephalosporins, or an aminoglycoside.

Preferred monotherapy for urosepsis due to enterococci (E faecalis) is with ampicillin or vancomycin (penicillin-allergic).

Empiric therapy for urosepsis due to unknown organisms is with piperacillin, imipenem, meropenem monotherapy or levofloxacin, aztreonam, or an aminoglycoside plus ampicillin.

Empiric therapy for other causes of sepsis

S aureus sepsis usually is associated with infection caused by devices or acute bacterial endocarditis (ABE). Empiric therapy may be with nafcillin, an anti-staphylococcal, cephalosporin, a carbapenem, linezolid, or clindamycin with or without rifampin.

Pneumococcal or meningococcal sepsis may be treated with penicillin G or a beta-lactam. If there is associated meningococcal meningitis, then the antibiotic selected should penetrate the cerebrospinal fluid and be given in meningeal doses.

Empiric therapy for sepsis of unknown origin

The usual sources of sepsis are from the distal gastrointestinal (GI) tract, pelvis, or GU tract. For IV-line infections, see above.

Organisms that should be covered from the GI/GU tract and pelvis include aerobic gram-negative bacilli (coliforms), B fragilis, and enterococci. (At least 90% of the organisms are E faecalis; not E faecium VRE.)

Preferred empiric monotherapy includes meropenem, imipenem, piperacillin/tazobactam, sulbactam/ampicillin, or cefoperazone.

	MISCELLANEOUS	Section 9 of 10
Author Information Introduction Clinical Differentials Workup Treatment Medication Follow-up Miscellaneous Bibliography

Medical/Legal Pitfalls:

• The most important medical/legal concerns regarding sepsis treatment are to ensure that the patient indeed does have sepsis, to rapidly identify its source, and to implement effective treatments. The most frequent error committed in treating patients who are septic is failure to consider pseudosepsis as a cause of the presenting syndrome complex. Most causes of pseudosepsis are readily treatable and reversible if recognized and treated early.

Special Concerns:

• Elderly patients may present with peritonitis and may not experience rebound tenderness of the abdomen. An acute surgical abdomen in a pregnant patient may be difficult to diagnose, but fortunately, most pregnant women are young, healthy, and physiologically strong. The most common cause of sepsis in pregnancy is urosepsis from an obstructed urinary tract, which may be due to the hormone effects of pregnancy on the ureters (hydroureters) and the mechanical obstructing effect of the uterus impinging upon the ureters.