SCCM Guidelines on COVID-19 Management – summarized

SCCM released a guideline on the management of COVID-19.  For those pressed for time, I’ve condensed the guidelines into a 13-page Q&A format.  The questions are formulated by me based on the guideline content and the answers are lifted from the SCCM guidelines, some of which I have reformatted from passive to active voice for easier reading.  Please reference the original guideline text, or message me if there are any errors.


(download PDF file here)


Surviving Sepsis Campaign: Guidelines on the Management of Critically Ill Adults with Coronavirus Disease 2019 (COVID-19). Waleed Alhazzani, Morten Hylander Møller, Yaseen M. Arabi , Mark Loeb, Michelle Ng Gong, Eddy Fan, Simon Oczkowski, Mitchell M. Levy, Lennie Derde, Amy Dzierba, Bin Du, Michael Aboodi, Hannah Wunsch, Maurizio Cecconi, Younsuck Koh, Daniel S. Chertow, Kathryn Maitland, Fayez Alshamsi, Emilie Belley-Cote, Massimiliano Greco, Matthew Laundy, Jill S. Morgan, Jozef Kesecioglu , Allison McGeer, Leonard Mermel, Manoj J. Mammen, Paul E. Alexander, Amy Arrington, John Centofanti, Giuseppe Citerio, Bandar Baw, Ziad A. Memish, Naomi Hammond, Frederick G. Hayden, Laura Evans, Andrew Rhodes

Hydroxychloroquine and Azithromycin for COVID-19

The group of Philippe Gautret (IHU-Méditerranée Infection, Marseille, France) presented their results on a non-randomized clinical trial using two antibacterial agents – hydroxychloroquine and azithromycin – for the treatment of COVID-19.

The results are promising, but the number of patients enrolled is small, and the study is not a randomized clinical trial.  Also, 6 of 26 patients in the treatment group were lost to follow-up, and no intention treat analysis has been included yet in the initial report.

The idea to use hydroxychloroquine, was based on a trial conducted in Chinese patients with COVID19 which showed a significant effect of chloroquine (an old antimalarial drug) on viral clearance as well as clinical outcome.  The dosing regimen for chloroquine was 500mg twice a day for ten days.

Hydroxychloroquine, which is an analogue of chloroquine, has been demonstrated to have anti-SARS-CoV activity in vitro, and has a better safety profile than chloroquine.

Azithromycin has been showed to be active in vitro against Zika and Ebola viruses, and prevent severe respiratory tract infections when given to patients with viral infection.

This study enrolled hospitalized patients older than 12 years old with documented SARS-CoV-2 carriage in nasopharyngeal sample on admission.  Patients with allergy or known contraindications to the drug (s.a. retinopathy, G6PD deficiency, QT prolongation) were excluded, as well as pregnant or breastfeeding patients.

The primary endpoint was virological clearance on day 6.

The treatment regimen is hydroxychloroquine sulfate PO 200 mg TID x 10 days.  The patients who refused treatment or were excluded based on criteria served as controls to the treatment group. Azithromycin was given to 6 patients (not part of the protocol) to prevent bacterial infection, at  a dose of 500mg PO x 1 then 250mg/day x 4 days.

Results of preliminary data showed that 70% of treatment group were virologicaly cured at day 6, compared with 12.5% in the control group (p = 0.001).  Subgroup analysis of patients who received both hydroxychloroquine and azithromycin showed that 100% of patients treated with both antibiotics were virologicaly cured, compared to 57.1% of patients treated with only hydroxychloroquine and 12.5% in the control group (p<0.001).

The study proposes that nasopharyngeal carriage of SARS-CoV2 can be cleared in 3-6 days in most patients.  Of note, mean duration of viral shedding (based on data from COVID cases in China) was 20 days (and up to 37 days).

Figure below:  percentage of patients with (+) nasopharyngeal samples through day 6.

Figure below:  percentage of patients with (+) nasopharyngeal samples through day 6.  Treatment group divided into 2 subgroups – hydroxychloroquine +/- azithromycin.


Raw data provided in the supplemental tables.  I colorized the table to show clearly why the data seems promising, with the limitations as discussed above.




Gautret et al. (2020) Hydroxychloroquine and azithromycin as a treatment of
COVID‐19: results of an open‐label non‐randomized clinical trial. International Journal of
Antimicrobial Agents – In Press 17 March 2020 – DOI : 10.1016/j.ijantimicag.2020.105949

Empiric Antibiotics for Brain Abscess

Base empiric therapy on presumed source of abscess and Gram stain results

Source: mouth, ear, sinus

  • [Penicillin (mouth) OR cefotaxime (ear/sinus)] + [metronidazole]

Source:  neurosurgery / post-op

  • [ceftazidime OR cefepime OR meropenem OR imipenem] + [vancomycin]

Source: penetrating trauma

  • [ceftriaxone or cefotaxime] + [vancomycin] +/- [metronidazole  (sinus involvement)]

Source:  hematogenous spread (IE, multiple abscess)

  • [ceftriaxone or cefotaxime] + [vancomycin] + [metronidazole]

Source: unknown

  • [ceftriaxone or cefotaxime] + [vancomycin] + [metronidazole]



  • Cefepime 2g IV q8h
  • Cefotaxime 2g IV q4-6h
  • Ceftriaxone 2g IV q12h
  • Ceftazidime 2g IV q8h
  • Imipenem 500-1000mg q6h
  • Meropenem 2g IV q8h
  • Metronidazole 15mg/kg IV load then 7.5mg/kg IV q8h;  usually 1G load & 500mg q8h
  • Nafcililn 2g IV q4h
  • Oxacillin 2g IV q4h
  • Penicillin G 20-24 M units / day IV in 6 divided doses
  • Vancomycin 15-20mg/Kg IV q8-12h



Uptodate: Treatment and prognosis of bacterial brain abscess, accessed 01/10/2017.


Antibiotic-Associated Encephalopathy

What are the clinical, radiologic, and electrophysiologic features of antibiotic-associated encephalopathy (AAE)?

3 types of AAE

  1. Type 1 AAE
    • onset within days of antibiotic initiation
    • common occurrence of myoclonus or seizures, abnormal EEG, normal MRI, and resolution within days
    • seen with penicillin and cephalosporins
    • most common in setting of renal insufficiency.
  2. Type 2 AAE
    • onset within days of antibiotic initiation
    • frequent occurrence of psychosis, rare seizures, infrequently abnormal EEG (nonspecific rather than epileptic), normal MRI, and resolution within days
    • seen with procaine penicillin, sulfonamides, fluoroquinolones, and macrolides
  3. Type 3 AAE
    • seen only with metronidazole
    • onset weeks after initiation
    • frequent occurrence of cerebellar dysfunction, rare seizures, rare and nonspecific EEG abnormalities, and omnipresence of abnormal MRI
  4. Isoniazid did not fit into any of these categories
    • time to onset is weeks to months
    • psychosis is common, seizures are rare, EEG is frequently abnormal but nonspecifically



*INH does not fit into any of the subtypes.



  • Time of onset after antibiotic initiation – 5 days (except INH and MNZ – 3 weeks)
  • Time to resolution of encephalopathy after ABx discontinuation – 5 days (except MNZ – 13 days)
  • MRI findings
    • abnormal in all MNZ-associated encephalopathy but normal in all others
      • T2 hyperintensities in dentate nuclei of cerebellum (also ?involvement of brainstem, corpus callosum, etc)
    • Bilateral frontal subcortical T2 MRI hyperintensities in isolated case of cefditoren pivoxil toxicity
  • CT findings
    • normal in all cases except 1 case of cerebellar hypodensity with MNZ toxicity and 1 report of L thalamic hypodensity with imipenem toxicity
  • EEG findings:
    • abnormal in 70%
    • abnormal in nearly all cases of cephalosporin-associated encephalopathy
    • common with PCN (83%), cipro (83%) and INH (69%)
    • most common abnormalities: nonspecific signs of encephalopathy (slowing, generalized periodic discharges with triphasic morphology
    • epileptiform discharges / seizures seen in 28% (55% in cephalosporins, 44% in quinolones and 40% of PCN, but not in macrolides, MNZ or sulfonamides)
  • check serum and CSF trough levels of antibiotics?
    • cefepime trough 0.2 to 1.1 mg/L
    • 50% probability of neurologic toxicity at trough of 22mg/L
    • CSF level during toxicity in 2 cases with values of 2.4 mg/L and 18 mg/L

A – FLAIR hyperintensities bilatearl deep cerebellar nuclei
B – DWI showing restricted diffusion in splenium with MNZ toxicity
C – ADC sequences






Bhattacharyya, Shamik et al. “Antibiotic-Associated Encephalopathy”. Neurology 86.10 (2016): 963-971.

Colonization vs. Infection

The cutoff for significant number of colony forming units to differentiate between colonization and infection depends on the diagnostic test:

  • tracheobronchial secretion, 10–5  CFU/ ml;
  • BAL, 10–4 CFU/ml
  • protected specimen brush, 10–3 CFU/ml



Bein, Thomas et al. “The Standard Of Care Of Patients With ARDS: Ventilatory Settings And Rescue Therapies For Refractory Hypoxemia”. Intensive Care Med 42.5 (2016): 699-711. Web. 14 May 2016.

Serratia marcescens

Serratia marcescens

  • anaerobic GNR, family Enterobacter
  • associated with hospital-onset infections
  • intrinsically resistant to ampicillin, amoxicillin, ampisulbactam, amoxiclav,, narrow-spectrum cephalosporins, cefuroxime,/ macrolides, tetracytclines, nitrofurantin and colistin
  • potential to harbor MDR mechanisms (AmpC or ESBL and carbapenemases)


  • uncomplicated infection – FQ, TMP SMZ, zosyn, ceph3 or ceph4, carbapenems
  • risk of AmpC-mediated resistance during therapy
    • likely with CNS infections, infections in sequestered sites that require prolonged antibiotic therapy, retained infected material
    • do not use ceph3 even if susceptible
    • in CNS infections – favor carbapenem or cefepime
    • other sequestered sites  FQ and TMP-SMZ
  • high level expression of AmpC beta-lactamase or ESBL – carbapenem

Duration of therapy 

  • depends on site of infection and clinical response
  • repeat culture/susceptibility testing and adjust Rx accordingly



Uptodate.  Serratia marcescens. Accessed 03/15/2016.

Intraventricular Antibiotics for Ventriculitis

Prepare the following:

1.  Three 3-way stopcocks

2.  One sterile saline flush (preservative-free)

3.  2-4 10cc syringes

4.  sterile gloves, sterile towels

5.  gauze with betadine

6.  cap, gown, mask

7.  antibiotic in 2cc syringe

Put drape underneath shunt access port.  Clean shunt access port with betadine thoroughly, paint line and port with betadine.  Prepare sterile field (won’t be completely sterile), put on gown, mask and sterile gloves.  Prepare stopcock, flush, empty syringe and antibiotic – connect in series as shown in photograph.  Maintain one hand as sterile and another hand as “dirty.”  Lock CSF drain to patient.  Connect free end of stopcock to shunt access port.  and open empty syringe (distal port) to patient.  Withdraw CSF into empty syringes – draw fluid slowly, to max of 20 cc. (volume equal to or slightly more than amount of antibiotic and sterile flush to be infused).  Close empty syringe (now filled with CSF) to patient.  Open antibiotic port (proximal port) to patient and push antibiotic slowly.  Close antibiotic port and open sterile flush port (middle port) to patient.  Flush enough saline to push antibiotic in tubing into patient, and then push an extra 1-2 ml more.  Close sterile flush port and disconnect intraventricular infusion set up from shunt access port.  Maintain EVD clamped x 1 hour.


Empiric treatment:

Vancomycin 15mgkg q8-12h (max 2g) plus one of the following

  1. ceftazidime 2g IV q8h
  2. cefepime 2g IV q8h
  3. meropenem 2g IV q8h

Gram positive:

  • Vancomycin for MRSA
  • Nafcillin or Oxacillin for MSSA
  • Add rifampin if refractory
  • Linezolid 600mg IV q12h if VRE or vancomycin allergy

Duration of treatment:

  1. normal CSF and CONS (+) – possible contaminatin, replace shunt on day 3 if cultures negative
  2. if CONS(+) and abnormal CSF – ABx while device in plus 1 week; document sterile CSF prior to shunt placement
  3. if virulent organism then >10d for staph or 14-21d for GNB; document sterile CSF x 10d prior to shunt
  4. if device cannot be removed, cont ABx until 7-10d after sterile CSF


  • best experience with gentamicin and vancomycin
  • may use colistin for MDROs (i.e. acinetobacter)
  • no PCN or cephalosporins (neurotoxic!)
  • goal is INHIB QUOTIENT of <10-20  … INHIB QUOTIENT is trough of CSF ABx / MIC

CHOICES: vanc 5-20mg/d; gent 4-8mg/d; ampho 0.1-1mg/d


  • IDSA recommends cefazolin but UPTODATE prefer vancomycin over cefazolin (pred CONS)
  • vancomycin 15mgkg (<2g) IV 2h prior (since vanc requires 60 minute infusion)
  • if al, then cefazolin 1-2g IV 1h prior



Intraventricular application of antibiotics to reach effective concentrations within the CNScapture


Very comprehensive review of intra-CSF antibiotics was published May, 2018 – author went over 200 articles on this topic – by Mrowczynski, et al published in Clinical Neurology and Neurosurgery.  See reference #3 below.  A short summary is provided here.

  1. Vancomycin
    • studies on intrathecal vancomycin for prophylaxis – potential use, 10mg/day x <10d
    • dosage ranges from 0.075 to 50mg/day, most common doses used 5-20mg/day
    • adverse effect: nerve root irritation in 1 case
  2. Teicoplanin
    • most commonly used for gram positive infections
    • most common doses used 5-20 mg/day, treatment duration 7-30 days
    • no significant side effects
  3. Daptomycin
    • useful for methicillin and vancomycin resistant strains
    • most common doses 2.5-10mg/day, treatment duration 1-4 weeks
    • adverse effect: transient pyrexia in 1 case, generally well tolerated
  4. Gentamicin
    • most common doses used 1-10mg/day, treatment duration 3-35 days
    • adverse effect: meningeal inflammation in 2 cases, mostly no significant side effects
  5. Amikacin
    • used at 4-50mg/day over 3 days to 6 months, ave duration 2 weeks
    • side effects:  hearing loss in 4 cases, tonic-clonic seizures in 1 case, vomiting in  1 case, back pain in 1 case
  6. Tobramycin
    • most often administered at 5-20mg/day x 2 infusions to every day for 40 days
    • no significant side effects
  7. Netilmicin
    • only 1 recent case report – 1mg to 150mg/day, no significant side effects
  8. Streptomycin
    • dosage varied from 2-100mg
    • associated with significant side effects – convulsions, apnea, coma, shock, pallor, vomiting, death
  9. Kanamycin
    • discontinued
    • used previously at dose of 5mg/day
  10. Penicillins
    • intrathecal therapy abandoned – multiple cases of severe seizures
  11. Lincomycin
    • older antibiotic used for gram positive infections
    • dose used 2mg/day x 5-9 days
  12. Cephalosporins
    • most common dosages used were between 25-100mg/d
    • note:  significant risk of seizures
  13. Erythromycin
    • varied from 3mg/day to 25mg/day, no significant side effects
  14. Polymyxin B
    • useful for multidrug resistant bacteria
    • most common dosages ranged from 2000 to 100,000 units x 5dto 4 weeks
    • one case described severeside effects: meningealirritation, decreased reflexes, but most reports noted no toxicity
  15. Colistin (Polymyxin E)
    • useful for resistant gram negative infections
    • predominantly used to treat acinetobacter baumanii infections at doses from 12,500 IU to 500,000 IU per day
    • Most common dose 125,000 IU
    • does not commonly cause toxicity, limited cases of chemical meningitis, nephrotoxicity, seizures and cauda equina syndrome noted
  16. Chloramphenicol
    • doses varied from 0.1mg to 50mg/day
    • no side effects noted
    • use can lead to aplastic anemia (intravenous) thus has been all but abandoned
  17. Rifampin
    • used at 2-5mg/day x 7-50 days
    • some patients have developed jaundice
  18. Isoniazid
    • dose range from 5-100mg/day, usually 3x per week
    • increased risk of significant side effects: hemiplegia, quadriplegia, convulsions, partial optic atrophy, hydrocephalus


Case report: successful use of intrathecal tigecycline after failed intravenous carbapenem antibiotic therapy of drug resistant Klebsiella ventriculitis.

Dose: 5mg/day, tigecycline diluted in saline up to a volume of 4ml; similar amount of CSF removed, and drug injected through EVD in 2 minutes; drain closed x 2 hours after dose; administered x total of 11d.

Pharma: 2 hours after dosing, levels peaked between 178.9 and 310.1 ug/mL, staying at 35.4-41.3 ug/mL at 6hours (which was 15-20 above MIC); after 24h, dose not detectable in CSF.

*reported doses in the literature = 1-10mg q12-24h; when los doses used, (+) CSF cultures persisted; IDSA recommends adjusting levels to 15-20x above MIC.


Nau, R., F. Sorgel, and H. Eiffert. “Penetration Of Drugs Through The Blood-Cerebrospinal Fluid/Blood-Brain Barrier For Treatment Of Central Nervous System Infections”. Clinical Microbiology Reviews 23.4 (2010): 858-883.

Mrowczynski, O., Langan, S. and Rizk, E. (2018). Intra-cerebrospinal fluid antibiotics to treat central nervous system infections: A review and update. Clinical Neurology and Neurosurgery, 170, pp.140-158.

Soto-Hernández, J., Soto-Ramírez, A., Pérez-Neri, I., Angeles-Morales, V., Cárdenas, G., & Barradas, V. (2019). Multidrug-resistant Klebsiella oxytoca ventriculitis, successfully treated with intraventricular tigecycline: A case report. Clinical Neurology And Neurosurgery, 188, 105592. doi: 10.1016/j.clineuro.2019.105592