A model approach

Larry H. Bernstein, MD

Executive Summary:

Sepsis is the most costly diagnosis in hospitalized patients and carries a high financial risk as a comorbidity and payment penalty under the new severity of illness CMS reimbursement guidelines as a patient safety hazard for failure to diagnose in a timely manner.  We carried out two studies of the early recognition of sepsis and related diseases in patients seen in the emergency department related to admission to the intensive care unit (ICU)(New YorkMethodistHospital) under the leadership of Lawrence Melniker, MD, Chairman of the Pharmacy and Therapeutics Committee.  The widely used SIRS criteria and the C-reactive protein, a long established acute phase protein, are each by themselves insufficient because of the low false negative rate of the former and the skewness and long tail of the latter related to uncurtailed noise from inconsequential inflammatory disease.  Using the elevated neutrophil count and left shift has proved to be elusive as well.  We and many others have established the validity of the European studies showing a marked benefit from using the procalcitonin (PCT, Brahms), and we propose to seize on the opportunity to calibrate the measurement of granulocyte maturation to the PCT.  The study would have to be carried out on a Sysmex instrument for accuracy and ease of use.  The Sysmex IG parameter is a measure of immature granulocyte counts and includes metamyelocytes, myelocytes and promyelocytes.

Background Study

Neutrophils thought to play a significant role in the early microvascular changes, are thought to be a key factor in the evolution of organ failure in the pathogenesis of severe sepsis and septic shock.  The mechanism of action of any drug or combination antibiotic combination therapy could potentially influence IG responses so that IG may be a useful way to monitor responses to therapy and disease progression by a simple, widely used hemocytometer that incorporates flow cytometry for cell identification.

An evaluation of the diagnostic performance of the Sysmex IG parameter and the procalcitonin (PCT, Brahms) assays when compared to existing practices and treatment decision guidelines is proposed following the establishment of a validated of a critical-decision cutoff for patients over 6 years old of 2.8% +0.2%.  Statistically significant numbers of samples representing the following patient groups would be assessed:

Group 1: Patients presenting to ICU with suspected Infection, SIRS (Systemic Inflammatory Response Syndrome, or severe sepsis, who are subjected to standard clinical and diagnostic investigation and do not fulfill criteria that warrant treatment with  antibiotic for the assessed state.

Group 2: Patients presenting to ICU with suspected Infection, SIRS (Systemic Inflammatory Response Syndrome, or severe sepsis, who are subjected to standard clinical and diagnostic investigation and are placed on antibiotic therapy (Infection) as a result of the investigation.

Group 3: Patients presenting to ICU with suspected Infection, SIRS (Systemic Inflammatory Response Syndrome, or severe sepsis, who are subjected to standard clinical and diagnostic investigation and are classified as having SIRS (Systemic Inflammatory Response Syndrome according to the equivalent of New York Methodist Hospital classification of SEPSIS (based on a modified Xigris (discontinue by Lilly) screening criteria, irrespective of the treatment option followed.

Group 4: Patients presenting to ICU with suspected Infection, SIRS (Systemic Inflammatory Response Syndrome, or severe sepsis, who are subjected to standard clinical and diagnostic investigation and have evidence of organ failure (New York Methodist Hospital classification of SEVERE SEPSIS; Xigris screening criteria for ACUTE ORGAN DYSFUNCTION.)

Diagnostic performance of both the Sysmex IG parameter and the procalcitonin (PCT, Brahms) assays are to be assessed as tools for sub-classification according to existing practices.

The ability of the IG parameter to detect the myeloid response associated prior to and increased with the onset of microvascular damage will be assessed. The potential to detect the IG response to predict progression towards multiple organ dysfunction could be an indication to initiate pharmacological therapy at a stage prior to significant evidence of organ failure.

The potential to use the IG parameter as a tool for monitoring responses to antibiotic and single or combination therapy could be assessed if the increase in IG shows good diagnostic performance alone or in combination as a necessary feature for decision-making.




GROUP 2: INFECTION–Does your patient have one or more of the following infection criteria?

  • Documented or Suspected–Does the patient have positive culture results (from blood, sputum, urine, etc.)?
  • Anti-Infective Therapy–Is the patient receiving antibiotic, antifungal, or other anti-infective therapy?
  • Pneumonia–Is there documentation of pneumonia (x-ray, etc.)?
  • WBCs–Have WBCs been found in normally sterile fl uid (urine, CSF, etc.)?
  • Perforated Viscus–Does the patient have perforated hollow organ (bowel)?
  • GROUP 3: SIRS-Does your patient have two or more of the following SIRS criteria?
  • Temperature–Is the patient’s temperature > 38°C (> 100.4°F) or < 36°C (< 96.8°F)?
  • Heart Rate–Is the patient’s heart rate > 90 bpm?
  • Respiratory Rate–Is the patient’s respiratory rate > 20 breaths/min?
  • WBC Count–Is the patient’s WBC count > 12,000/mm3, < 4000/mm3, absolute neutrophil count > 11,000/mm3, or are there > 2.8% immature granulocytes (myelocytes and metamyelocytes) discounting 10% band neutrophils and the less mature promyelocytes for left shift?
  • GROUP 4: ACUTE ORGAN DYSFUNCTION-Does your patient have one or more of the following organ dysfunction critera?
  • Cardiovascular–Does the patient have a systolic BP ≤ 90 mmHg or mean arterial pressure ≤ 70 mmHg (for at least 1 hour despite fl uid resuscitation) or require vasopressor support?
  • Respiratory–Does the patient have a PaO2/FiO2 ratio ≤ 250, PEEP > 7.5 or require mechanical ventilation?
  • Renal–Does the patient have low urine output (eg, <0.5 mL/kg/hr for 1 hour despite adequate fl uid rescuscitation),
  • increased creatinine (>50% increase from baseline) or require acute dialysis?
  • Hematologic–Does the patient have a low platelet count (< 100,000/mm3) or PT/PTT > upper limit of normal?
  • Metabolic–Does the patient have a low pH with high lactate (eg, pH < 7.30 and plasma lactate > upper limit of normal?
  • Hepatic–Are the patient’s liver enzymes > 2x upper limit of normal?
  • CNS-Does the patient have altered consciousness or reduced Glasgow Coma Score?

Guidelines For Management

Sepsis, Severe Sepsis, and Septic Shock

A. Definitions


Presence or Suspicion of infection and one or more of the following conditions

  • Fever (core temperature >38.3°C)
  • Hypothermia (core temperature <36°C)
  • Heart rate >90/min or >2 SD above the normal value for age
  • Tachypnea > 20/min or >2 SD above the normal value for age
  • Altered mental status
  • Leukocytosis (WBC count >12,000/µL)
  • Leukopenia (WBC count <4000/µL)
  • Neutrophilia as defined above
  • Normal WBC count with >2.8% immature granulocytes (IG)


Severe Sepsis

Sepsis and at least one New Organ Dysfunction

Organ dysfunction variables:

  • Altered level of consciousness or reduced Glasgow coma score
  • Arterial hypoxemia (PaO2/FIO2 <300)
  • Acute oliguria – urine output <0.5 mL/kg/hr)
  • Creatinine > 2.0 mg/dL or > 50% increase from baseline
  • Coagulation abnormalities (INR >1.5 or aPTT >60 secs)
  • Thrombocytopenia (Platelet count <100,000/µL)
  • Hyperbilirubinemia (Plasma total bilirubin > 2.0 mg/dL or 35 mmol/L)


Tissue perfusion variables:

  • Hyperlactatemia (>2 mmol/L)
  • Metabolic acidosis  ( pH < 7.30)


Hemodynamic variables:

  • Transient arterial hypotension (SBP <90, MAP <70, or SBP decrease >40 mm Hg from baseline) (Hypotension corrected with adequate volume resuscitation)


Septic Shock

Severe Sepsis and Persistent Arterial Hypotension

Screening Tool for Sepsis

Emergency Department, Med-Surg Floors, and Critical Care Units


1. Is the patient’s history suggestive of a NEW infection?                                                             ___ Yes ___No


[Check any that apply]

Pneumonia or Empyema

(  )

Skin/soft tissue infection

(  )

Urinary tract infection

(  )

Wound infection

(  )

Acute abdominal infection

(  )

Bone/joint infection

(  )


(  )

Bloodstream catheter

(  )


(  )

Implantable device

(  )


(  )



2. Are any two of the following signs, symptoms, or findings of infection

*both* – Present and New – to the patient?                                                                                    ___ Yes ___No


[Check any that apply]


> 38.3 °C (101.0 oF)


< 36 °C     (96.8°F)

(  )

(  )


(WBC count >12,000/µL)


(WBC count <4000/µL)

(  )

(  )

Tachycardia > 90 bpm

Tachypnea > 20 bpm

Altered mental status

(  )

(  )

(  )


(serum glucose >120 mg/dL

– in the absence of diabetes)

(  )

If The Answer Is “YES” To BOTH Questions 1 And 2,




  • Immediately obtain:
    • CBC with differential
    • Comprehensive metabolic panel
    • Procalcitonin
    • C-reative protein (CRP)
    • Lactate level
    • ABG
    • Blood cultures
    • Liver function tests
    • Coagulation profile
    • Urine analysis
    • CXR
    • Pulse co-oximetry

3. Are any of the following organ dysfunctioncriteria *both* – Present & New – in an organ remote from the site of the infection? 

___ Yes ___No



Organ Dysfunction Criteria


  • SBP < 90 mmHg or MAP < 65 mmHg
  • SBP decrease > 40 mm Hg from baseline
  • Bilateral pulmonary infiltrates with a:
New or increased O2 supplementation requirement to maintain SpO2 > 90%   OR
  • PaO2/FiO2 ratio < 300
  • Creatinine > 2.0 mg/dl (176.8 mmol/L)
  • Urine Output < 0.5 ml/kg/hour for > 2 hours
  • Bilirubin > 2 mg/dl (34.2 mmol/L)
  • Platelet count < 100,000
  • Coagulopathy (INR >1.5 or aPTT >60 secs)
  • Lactate > 2 mmol/L (18.0 mg/dl)

Note: the remote organ stipulation is waived in the case of bilateral pulmonary infiltrates


If suspicion of infection AND organ dysfunctionare present,

the patient meets the criteria for SEVERE SEPSIS


èInitiate severe sepsis protocol to achieve these goals <6 hrs



6-Hr Goals for Severe Sepsis
1)    Mean arterial pressure > 65 mm of Hg
2)    Urine output > 0.5 ml/kg/hr [average sized adult > 30-40 cc/hr]
3)    CVP > 8-10 mm Hg or Sonographic Signs of adequate filling pressures
4)    SVO2 > 70%

Early Goal-Directed Therapy for Severe Sepsis

[For Emergency Department / Med-Surg Floors / Critical Care Medicine settings]


First 6 hrs


  • Severe sepsis identified by screening                 Yes                  No
  • Blood cultures sent                                           Yes                  No
  • Serum lactate sent                                             Yes                 No
  • Patient hypotensive with
    • Systolic Blood Pressure <90 or                   Yes                  No
    • Mean Arterial Pressures < 65 mm of Hg     Yes                  No

0 – 1 hr Management

1)      Start fluid bolus normal saline 20 ml/kg at 500 – 1000ml over 30 minutes and re-evaluate blood pressure and urine output (expected value >0.5ml/kg/hr)

2)      Re-evaluate 10 minutes after fluid bolus

3)      If blood pressure is stabilized, continue fluids at maintenance rate [No CVP Monitoring Needed]

4)      O2 supplementation to maintain SaO2 > 90% — ventilatory support, if indicated

1 – 2 hrs Management

5)      If patient remains hypotensive [Med-Surg MUST Call for CVP Monitoring Approval]

  • ABG, if not done already
  • Measure CVP or Sonographic Signs of adequate filling pressure:                                           When possible:
    • Central Venous Catheterization with
    • Central Venous Pressure Transducer/Monitor



For CVP < 8-10 mm Hg and MAP < 65 mm Hg


  • Repeat fluid bolus 20 ml/kg at 500 –1000 ml over 30 minutes until:
    • Patient has CVP > 8-10 mm Hg
      • Continue fluid boluses to correct CVP > 8-10 mm Hg


  • Signs of volume overload on physical examination
  • If patient is unstable
    • May start norepinephrine infusion at this time

For CVP > 8-10 mm Hg and MAP < 65 mm Hg


  • Start norepinephrine infusion to achieve MAP >65 mm Hg


1 – 2 hrs Management (continued)


6)      Stat antibiotics (suggested agents – adjust for creatinine clearance)


  1. a.      Community Acquired Pneumonia  – Follow hospital protocol
  2. b.      Healthcare Associated Pneumonia – Follow hospital protocol
  3. Urinary tract infection (choose one)
    1. Ceftriaxone 1 gm IVPB (Community Acquired)

ii. Cefipime 1 gm IVPB (Hospital Acquired)

  1. Ciprofloxacin 400 mg IVPB (For PCN or Cephalosporin allergy)
  2. Suspected intra-abdominal infection (choose one)
    1. Cefipime 1 gm IVPB and Metronidazole 500 mg IVPB

ii. Ciprofloxacin 400 mg IVPB and Metronidazole 500 mg IVPB

  1. Piperacillin /Tazobactam 3.75 gm IVPB



2 – 6 hrs Management

6) Admit/Transfer the patient to Critical Care Medicine setting

7) Repeat lactate level in 4 hrs, if lactate > 4 mmol /L:                                                      Measure SVO2

8)  If SVO2 < 70 % & HCT < 30%:                                                                               Consider transfusion of PRBC to achieve HCT > 30%

9) Repeat SVO2 after optimization of CVP > 8-10 mm Hg & HCT > 30%:         Consider inotropic therapy

10) If SVO2 remains < 70%:                                                                                           Start Dobutamine infusion at 5 micrograms/kg/minute

6 –24 hrs Management [Critical Care Medicine]


   11) Evaluate for Relative Adrenal Insufficiency:                                                                      Send serum cortisol level and order cosyntropin test

12)While waiting for cortisol level:                                                                                   Start Decadron 4 mg IV

13) For blood sugar > 150 mg/dl:                                                                                                Start Critical Care Medicine Insulin protocol

14) Evaluate for drotrecogin alpha administration:                                                          Use Hospital protocol

15) If patient is on a ventilator, evaluate for ALI/ARDS:                                                  Start ARDS ventilator protocol in appropriative patients

16 If patient remain hypotensive and CVP > 8-10 mm Hg:                                                           Start Vasopressin infusion at 0.04 units per minute

Summary Approach to Problem

Objective: To sub-classify patients presenting to ICU into GROUPS 1-4 as described above, and to record the appropriate treatment decision (Antibiotics, and other).  Statistically significant numbers of patients representing each of the sub-groups will be included in the study.  Diagnostic performance of both the Sysmex IG parameter and the procalcitonin (PCT, Brahms) assays will be assessed as potential tools to differentiate groups 1 from 2, 2 from 3 and 3 from 4.

If the IG and/or procalcitonin tests are deemed valuable as markers of the microvascular damage which precedes multiple organ damage associated with sepsis, the potential exists to motivate for application as an index for the initiation of innovative drug therapy at an earlier stage, in an effort to prevent disease progression to multiple organ failure.  If data supports this change, the potential for monitoring responses to investigated antibiotic therapy in patients with raised IG and / or PCT values should be assessed.

Methods: A prospective, observational study from one or several large community or academically-linked hospitals following IRB requirements. A total of 1000 consecutive patients presenting to ICU with presumed infection/sepsis will be enrolled.  Clinical and diagnostic sub-classification according to groups 1-4 above have to be performed, in conjunction with a Sysmex CBC, Diff and IG, as well as a procalcitonin (PCT, Brahms) assay.

Results:  Statistically significant numbers of patients representing each of the 4 groups will be documented, and their treatment (antibiotic/combination drug) will be recorded.

Statistical assessment: At a minimum, ROC curve analysis of IG (and PCT) versus Group 1-4 classifications will be done. ROC curve analysis of IG (and PCT) versus therapy decisions will also be performed.  If successful, the capability of the parameters to monitor treatment responses will be assessed by serial measurements over time. A method of anomaly characterization developed by Gil David and Prof. Ronald Coifman of Yale University will be applied using key indicators to classify the patients such as WBC, percent neutrophils, IG, PCT, subclass 1-4, treatment, outcome (LOS in ICU, LOS, died).

Study Design:
Type study: Prospective and not interventional
Patient population: 1000 patients, admitted to ICU with suspected severe infection / SIRS / Sepsis.

Diagnostic information: Concurrent information gathered will be in accordance with Xigris Screening recommendations for severe sepsis, and theHospital guidelines for management of sepsis, severe sepsis and septic shock as described above.  Including – respiratory rate, heart rate, fever, location, primary and secondary diagnoses, APACHE score and SOFA score, antibiotic use, target therapy/other use etc.


About larryhbern

I currently have been engaged engaged with a medical informatics project called "Second Opinion" with Gil David and Prof. Ronald Coifman in the Program in Applied Mathematics at Yale. We plan a proof of concept study. I was the Chief, Clinical Pathology at NY Methodist Hospital, a 600+ bed hospital in Park Slope, Brooklyn, 2 hours from Bridgeport, CT, where I worked for 5 years, and was previously Chief of Clinical Chemistry and Chief of Blood Bank at Bridgeport Hospital for 22 years, and Acting Chairman of Yale University Department of Pathology at Bridgeport Hospital for one year prior to my experience at NY Methodist Hospital Weill-Cornell. I have done extensive work with nutrition (co-chaired the First International Transthyretin Congress in Strasbourg, chaired the 14th Ross Roundtable in Nutrition and the Beckman Roundtable on Prealbumin, and was recipient of the Labbe/Garry award of the Nutrition Division of AACC), troponins and with NT-proBNP and have current projects in normalizing the NT-proBNP for age and estimated glomerular filtration rate, with good success. I have served on the Board of Directors of NAACLS and the American Library Association Commission on Accreditation, am listed in America’s Top Physicians, Marquis Who’s Who in Science and Engineering and Marquis’ Who’s Who in Medicine.
This entry was posted in bacteremia, sepsis, sepsis biomarkers, septic shock, SIRS and tagged , , , , , , , , , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s