Healthcare-associated Sepsis in the Neonatal Intensive Care Unit: A Practical Approach

BACKGROUND

Healthcare-associated (HCA) sepsis (known as late-onset/nosocomial sepsis) is one of the leading causes of mortality and morbidity in neonatal intensive care units (NICU). Approximately 25% of premature neonates develop HCA sepsis during NICU stay based on published data. This figure is probably higher in limited resource countries. Every effort should be directed toward prevention rather than treatment of HCA sepsis. This article gives a practical approach to HCA sepsis and beyond the scope of pathophysiology, causative agents, risk factors and other aspects of classical review article.

PURPOSE

Diagnosis and management of HCA sepsis for infants in NICU.

CLINICAL MANIFESTATIONS

The clinical symptoms and signs of neonatal sepsis are diverse and nonspecific and may occur in the absence of sepsis. Various symptoms and signs suggestive of sepsis include recurrent apnea, tachypnea, tachycardia, hemodynamic instability, cyanotic spells, temperature instability, mottled skin, poor feeding, feeding intolerance, decreased activity, irritability, seizure and lethargy. Occasional episodes are discovered incidentally when blood work shows an abnormality such as leucocytosis, leucopenia, thrombocytopenia or hyperglycemia. Focal signs such as abscess, swollen joints or bones, pseudoparalysis or erythema can be the only finding to suggest sepsis. More recently, changes in heart rate such as loss of beat to beat variability and significant deceleration have been attributed to development of sepsis.

DIAGNOSIS

Complete sepsis workup

Blood culture

Owing to high colony count bacteremia in neonatal sepsis a minimum of 1 cc of blood is generally sufficient. Blood should be obtained after careful skin antisepsis and sent for aerobic culture prior to starting antibiotic therapy. Anerobic cultures provide limited additional benefit. Additional blood culture (BC) should be withdrawn through central venous catheter if one exists.

Lumbar puncture

Lumbar puncture (LP) should be done “if no contraindications” and cerebrospinal fluid (CSF) should be sent for culture, Gram stain, chemistry and cell count. Additional sample can be saved for viral study if clinically feasible. If there is a contraindication, LP should be deferred until the patient is more stable.

Normal CSF values

Several studies have shown a wide range of CSF values that differ according to gestational age and postnatal age.

Cell count of 30 cells/mm³ has been used conventionally as a cut-off for normal value while the studies indicate it may range from 0 to 42 cells/mm³. It is more important to look at the differential than the absolute cell number. Lymphocytes and monocytes are predominant in the CSF under normal circumstances.

It is important to emphasize that up to 15% of culture proven neonatal meningitis may show normal CSF cell count in the initial evaluation.

CSF protein value is quite variable as well; it may range from 30 to 130 mg/dl in term neonates. In premature neonates the upper limit may exceed 130 mg/dl.

CSF glucose should be compared with serum glucose, and it should be above two-thirds of the serum level.

Urine analysis and culture

Clean urine sample (suprapubic aspiration or catheter) should be sent for analysis and culture.

Other sites’ culture

Other possible foci for sepsis may include: skin lesions, abscess, wound, tracheal aspirate (If clinically indicated)

Complete blood count and differential

Many changes in complete blood count have been observed during neonatal sepsis episodes: leucocytosis, leucopenia, neutropenia, increase immature–to-total neutrophil (I:T) ratio above 0.2 and thrombocytopenia.

I:T ratio has the best negative predictive value among all parameters.

No single measurement or value has good sensitivity or positive predictive value. So, a normal study should not provide reassurance to defer treatment in clinically septic neonates.

Adjuvant tools

C-reactive protein and other acute phase reactants

C-reactive protein (CRP) is one of the adjuvant tools for sepsis diagnosis; it has better sensitivity if obtained at 20-24 hrs after sepsis onset. Serial levels of CRP are helpful in monitoring the effect of antibiotic therapy.

Other new acute phase reactants such as procalcitonin, interleukins (6, 8, 1-β) and tumor necrosis factor-α and many others, have better sensitivity than CRP and are being used in many centers around the world.

Viral studies

Culture, PCR or serology as clinically indicated. Disseminated Herpes simplex infection may mimic bacterial sepsis.

Others

Chest radiograph, ultrasonography, echocardiography may help to localize a focus of infection.

MANAGEMENT

Antimicrobial chemotherapy

Empiric antibiotic choice should be guided by the local antibiogram data for each medical institute. The guidelines for antibiotic choice are designed to minimize overuse of extended spectrum antibiotics and to limit emergence of multidrug resistant organisms (MDR). Usage of third generation Cephalosporins especially ceftazidime, carbapenems and vancomycin should be restricted.

Stable infant

Empiric therapy with cloxacillin and an aminoglycoside (gentamicin or amikacin) is a widely acceptable choice. Amikacin has better Gram-negative coverage especially in the NICU where multidrug resistant or ESBL* organisms are endemic. If the patient has central line in situ, vancomycin might be preferable for the initial coverage instead of cloxacillin.

Unstable infant

In the critically ill neonate with hemodynamic instability, severe acidosis, etc., broader spectrum antibiotics (such as: cefotaxime and vancomycin) should be considered after discussion with neonatology or ID services.

• Antibiotic coverage must be changed according to the reported sensitivity pattern of the isolate, to an effective drug with the narrowest spectrum.

• Antibiotic serum level of vancomycin or aminoglycoside should be obtained according to routine protocol or even earlier if there is renal impairment.

Ongoing evaluation

If the initial cultures are negative at 48-72 hours

If the infant is stable, CRP is normal and the likelihood of sepsis is low or the condition was explained by another cause, antibiotics can be stopped and the patient observed.

If the patient's condition unchanged, consider repeating blood culture, searching for other causes and re-evaluating the antibiotic coverage.

If the initial cultures are positive

If coagulase-negative Staphylococcus (CoNS) is isolated from a blood culture and the patient's condition is improving and platelet level is within normal range continue the antibiotics unchanged (i.e., cloxacillin or vancomycin) for 7-10 days.

In case of thrombocytopenia with CoNS, a repeat blood culture should be obtained, Vancomycin should be started and other antibiotics should be discontinued.

In case of persistent (two or more positive blood culture) CoNS sepsis, rifampicin can be added to vancomycin.

If Gram-negative organisms, Staphylococcus aureus or yeast were isolated from blood or CSF, repeat the cultures no later than 48 hours after adjusting antibiotics.

If the initial antibiotic coverage is not appropriate based on the culture results:

Repeat the culture immediately and change to appropriate coverage and repeat the culture after 48 hours.

No need to repeat urine culture in general.

Duration of therapy is location and organism dependent [Figure 1].

Figure 1.

HCA sepsis algorithm; ID - Infectious disease; GNR - Gram negative rods; GBS - Group B Strept; SA - Staphylococcus aureus; Plt - Platelets; d/c: discontinue; Clox - Cloxacillin, ESBL - Extended spectrum beta-lactamase

ID consultation if not already requested.

Persistence of sepsis

Search for the focus: lines, endovascular infection, abscess, fungal ball, etc.

Reevaluate antibiotic coverage, dosing, resistance, etc.

Line removal is advised whenever possible.

NB: it is expected for fungemia to persist for several (7-10) days.

Adjuvant therapy

• Intravenous Immunoglobulin and G-CSF use are still debatable and probably not beneficial

• Exchange transfusion: no evidence to support its use in neonatal sepsis and should be discouraged.

FOR FURTHER READING

• 1.Stoll BJ, Hansen N, Fanaroff AA, Wright LL, Carlo WA, Ehrenkranz RA, et al. late-Onset sepsis in very low birth weight neonates: The experience of the NICHD neonatal research network. Pediatrics 2002;110:285-91.
• 2.Lawrence SL, Roth V, Slinger R, Toye B, Gaboury I, Lemyre B. Cloxacillin versus vancomycin for presumed late-onset sepsis in the Neonatal Intensive Care Unit and the impact upon outcome of coagulase negative staphylococcal bacteremia: A retrospective cohort study. BMC Pediatr 2005;5:49.
• 3.Blayney MP, Al Madani M. Coagulase-negative staphylococcal infections in a neonatal intensive care unit: In vivo response to cloxacillin. Paediatr Child Health 2006;11:659-63.
• 4.Guerti K, Devos H, Ieven MM, Mahieu LM. Time to positivity of neonatal blood cultures: Fast and furious? J Med Microbiol 2011;60:446-53.
• 5.Berner R, Niemeyer CM, Leititis JU, Funke A, Schwab C, Rau U, et al. Plasma levels and gene expressionof granulocyte colonystimulating factor, tumor necrosis factor-alpha, interleukin (IL)-1beta, IL-6, IL-8, and soluble intercellular adhesion molecule-1 in neonatal early-onset sepsis. Pediatr Res 1998;44:469-77.
• 6.Bomela HN, Ballot DE, Cory BJ, Cooper PA. Use of C-reactive protein to guide duration of empiric antibiotic therapy in suspected early neonatal sepsis. Pediatr Infect Dis J 2000;19:531-5.
• 7.Lopez Sastre JB, Solis DP, Serradilla VR, Colomer BF, Cotallo GD. Evaluation of procalcitonin for diagnosis of neonatal sepsis of vertical transmission. BMC Pediatr 2007;7:9.