Pneumonia, ventilator-associated (VAP)

From CCMDB Wiki
Jump to navigation Jump to search
ICD10 Diagnosis
Dx: Pneumonia, ventilator-associated (VAP)
ICD10 code: J95.88
Pre-ICD10 counterpart: VAP - Ventilator Associated Pneumonia
Charlson/ALERT Scale: none
APACHE Como Component: none
APACHE Acute Component: none
Start Date:
Stop Date:
External ICD10 Documentation

This diagnosis is a part of ICD10 collection.

  • SMW
    • 2019-01-01
    • 2999-12-31
    • J95.88
  • Cargo

  • Categories
  • SMW
  • Cargo

  • Categories
  • SMW
  • Cargo

  • Categories
  • SMW
  • Cargo

  • Categories
  • SMW
  • Cargo

  • Categories
  • SMW
  • Cargo

  • Categories
  • SMW
  • Cargo

  • Categories

Additional Info

Regarding the Date of Onset

  • The incident date for a VAP should be at the onset of the infection or when you determine the VAP was first brewing.

This is determined when reviewing the chart retrospectively. See Dr. Garland's comments below:

  • In general we want the date on which the VAP was first evident -- in retrospect. This MAY NOT BE THE DAY it was first recognized as being present in real time by the medical team.
    • e.g. An intubated patient had a CXR on Thursday showing a little wispy infiltrate on the CXR. In the absence of other signs or symptoms, on that day the team did NOT think it was infectious. But Friday the patient developed fever and leukocytosis and purulent sputum, AND the wispy infiltrate was now a big, dense consolidation. A sputum culture was sent on Friday for the first time. At this point the team began antibiotics for pneumonia. The thing here is that only in RETROSPECT did it become clear that the wispy infiltrate seen on Thursday WAS the start of the VAP. Thus, in this case the VAP appears to have clinically begun on Thursday, not Friday.
      • NOTE that IF the intubation was Tuesday or Wednesday or Thursday, then this is NOT a VAP, because the clinical onset of the pneumonia was <48 hours prior to intubation. If the intubation was Monday or prior, then it is a VAP.
  • This issue of timing can be VERY tricky -- and will always require judgement and retrospective assessment of the sequence of events.
    • e.g. Patient has had fever and leukocytosis for 5 days due to a septic gallbladder, and has been intubated that whole time. Now a new infiltrate with shows up, with purulent sputum and the team believes a new pneumonia has developed. So here you can't use the pre-existing fever and elevated WBC to identify the clinical onset, and it's the change in the CXR that makes it.
    • e.g. Patient has ARDS from multiple trauma and so the CXR has had diffuse fluffy infiltrates for a week. He's also had a low-grade fever the whole time. Now the fever becomes high-grade, the sputum becomes purulent, and though it's hard to tell for sure, the CXR seems to be a bit worse in the RUL. The team concludes a pneumonia has developed. So here, it's a judgement that the subtle change in the CXR and the change in the fever curve and the change in sputum is due to a VAP.

Data Collection Instructions

VAP Criteria

1 - Ventilated 48 hrs

VAP is an infectious pneumonia in a patient who, as of the day it showed itself (“day of event”) had been on mechanical ventilation (MV), either continuously or intermittently for at least 48 hours before onset of infection.

  • The mechanical ventilation must be delivered via an endotracheal tube or tracheostomy.
  • Although it is arbitrary, for this purpose we will consider "intermittent ventilation" to mean this: Over the 48 hours prior to the identification of the VAP, that the patient had been on the ventilator, via an ETT or trach, at least twice for periods of at least 1 hour each.
  • Regarding whether ALL evidence/symptoms/signs of the pneumonia must be absent for the 48 hours after intubation for it to be called a VAP:
    • The short answer is No for Criterion#2 but Yes for the Criteria#3,4 and 5, i.e. the respiratory symptoms/signs
    • Example 1: The patient has had a pyelonephritis with leukocytosis and fever for the past 6 days. He got intubated 4 days ago and developed new respiratory signs and symptoms today, i.e. 72 hours after intubation. So even though Criterion#2 was present before the intubation and >48 hrs ago, the onset of the respiratory signs/symptoms were not present >48 hrs before intubation --- this IS a VAP.
    • Example 2: The patient got intubated 4 days ago and developed new fever and leukocytosis yesterday (3 days after intubation). And while he has the required respiratory symptoms/signs, they developed just 24 hours after intubation. This is NOT a VAP, as the respiratory findings began <48 hrs after intubation.

2 - Infection

If not immunocomprised

  • has at least ONE of the following 3 things:
    • Fever > 38.0
    • WBC<4000 or >12,000
    • If >70 years old, altered mental status without another recognized cause
  • Note that these symptoms/signs can be present without infection, but it's often difficult to tell which they're from. So if there is a VERY high level of belief that the fever or WBC changes are NOT due to infection, then don't count such criteria as being present. But in the absence of such high certainty then DO count them toward the VAP diagnosis. It's a judgement call.

If Immunocompromised

  • There are different criteria for immunocompromised patients
    • immunocompromised for this purpose defined as any of:
      • neutropenia defined as absolute neutrophil count or total white blood cell count (WBC) <500/mm3
      • leukemia, lymphoma or who are HIV positive with CD4 count <200
      • have had a splenectomy
      • history of solid organ or hematopoietic stem cell transplant
      • on cytotoxic chemotherapy
      • on steroids (excluding inhaled steroids) daily for >2 weeks on the date of event
    • Immunocompromised, must have at least ONE of the following 8 things:
      • Fever (>38.0°C)
      • If >70 years old, altered mental status without another recognized cause
      • New onset of purulent sputum, or change in character ofsputum, or increased respiratory secretions, or increased suctioning requirements
      • New onset or worsening cough, or dyspnea, or tachypnea
      • Rales or bronchial breath sounds
      • Worsening gas exchange (for example: O2 desaturations, increased oxygen requirements, or increased ventilator demand)
      • Hemoptysis
      • Pleuritic chest pain

3 - CXR indicators

  • Chest imaging (X-ray) study or studies showing at least ONE of the following 3 things, that must be new & persistent OR progressive and persistent: (CDC WORDING CHANGE).
    • Infiltrate -- note that there are alternative words used for infiltrates, including "airspace opacities"
    • Consolidation
    • Cavitation
  • Regarding the identification of "persistent" here:
    • Because it is well recognized that there are many non-infectious reasons for fleeting infiltrates in intubated ICU patients (e.g. atelectasis), #3 - CXR indicators requires that the listed changes be either "New and Persistent" or "Progressive and Persistent". This necessarily means that there must be >1 CXR -- preferably over more than 1 day -- showing the "persistence".
    • Although it is somewhat arbitrary, we will consider chest imaging findings to be "persistent" if they remain similar (i.e. do not disappear, or mostly disappear) in imaging studies done at least 6 hours apart.
    • But, of course, ICU patients who are believed to have significant lung pathology, including pneumonia, typically DO have followup CXRs that will allow for identification of persistence of the changes seen.
    • In the (relatively rare situation) in which a ventilated patient qualifies for a VAP except that NO CXR was done during the next couple of days to demonstrate persistence, you could point out to the physician(s) that the CDC criterion require infiltrates be persistent by chest imaging and therefore we would require a followup CXR to confirm the diagnosis. If there is severe resistance to this from the ICU team, you could refer them to Drs. Garland or Paunovic.


EITHER of #4 (respiratory indicators) OR #5 (alternative indicators)

4 - Respiratory indicators

Has at least TWO of the following 4 things:

  • New onset of purulent sputum or change in character of sputum, or increased respiratory secretions, or increased suctioning requirements.
  • New onset or worsening cough, or dyspnea, or tachypnea
  • Newly identified adventitia, crackles or bronchial breath sounds.
  • Worsening gas exchange -- e.g., O2 desaturations (e.g., PaO2/FiO2 <240, arterial blood gas or pulse oximetry), increased oxygen requirements, or increased ventilator demand


5 - Alternative indicators

Has at least ONE item in #List 5a AND at least ONE item in #List 5b

List 5a

(This is the same list as #4 - Respiratory indicators, however, here it's just 1 item instead of 2 items)

  • New onset of purulent sputum or change in character of sputum, or increased respiratory secretions, or increased suctioning requirements.
  • New onset or worsening cough, or dyspnea, or tachypnea
  • Newly identified Rales or bronchial breath sounds.
  • Worsening gas exchange -- e.g., O2 desaturations (e.g., PaO2/FiO2 <240, arterial blood gas or pulse oximetry), increased oxygen requirements, or increased ventilator demand.


List 5b

  • Organism identified from blood (see #VAP Pathogen exclusion list)
  • Organism identified from pleural fluid (see #VAP Pathogen exclusion list)
  • Virus, Bordetella, Legionella, Chlamydia or Mycoplasma identified from respiratory secretions or tissue by a culture or non-culture based microbiologic testing method
  • In an immunocompromised patient: identification of matching Candida from blood and sputum, endotracheal aspirate, bronchoalveolar lavage (BAL) or protected specimen brushing. (This is a new CDC item)
  • In an immunocompromised patient: Evidence of fungi (i.e. mycelia, not yeast) from BAL or protected specimen brushing)(This is a new CDC item). Be aware that fungi is not the same as yeast (do not use candida spp).
  • Positive quantitative culture, performed according to accepted protocols, from BAL or protected brush specimens (see#VAP Pathogen exclusion list)
    • According to Dr. Garland this very rare and he has never seen one done at HSC. Most bronchoscopies are not "quantitative" therefore not valid for use in the VAP criteria unless candida is cultured in immunocompromised patients only and only if the patient also has a matching blood culture.
  • >5% of cells obtained from BAL contain intracellular bacteria on direct microscopic exam
  • Positive culture of lung tissue (see #VAP Pathogen exclusion list)
  • Histopathologic exam of lung tissue identifies abscess formation, or foci of consolidation with intense PMN accumulation in bronchioles and alveoli
  • Histopathologic exam of lung tissue identifies lung invasion of fungal hyphae or pseudohyphae
  • Fourfold rise in paired sera (IgG) for pathogen (e.g., influenza viruses, Chlamydia)
  • Fourfold rise in Legionella pneumophila serogroup 1 antibody titer to ≥1:128 in paired acute and convalescent sera by indirect IFA.
    • Detection of L. pneumophila serogroup 1 antigens in urine by RIA or EIA

VAP Pathogen exclusion list

  • NEW CDC LIST (does not include candida b/c can be a valid pathogen in immunocompromised pts who have candida in both blood and ETT cultures (see list 5B, above)
    • Normal respiratory flora
    • Normal oral flora
    • Mixed respiratory flora
    • Cagulase-negative staph species (includes S. epidermidis, does not include S. aureus)
    • Enterococcus species
    • Blastomyces species (blasto)
    • Histoplasma species
    • Coccidioides species
    • Paracoccidioides species
    • Cryptococcus species
    • Pneumocystis species
  • Patients might be treated for infection with these pathogens, but do not code them as VAP. In that case you might be able to code it as a Hospital-acquired pneumonia (HAP) in ICD10 or Community-acquired pneumonia (CAP) in ICD10.
  • Being Colonized with organism (not infected) with MRSA does not exclude it from causing VAP, if they meet the listed criteria.

Sputum culture results do not qualify for VAP

Endotracheal tube secretion / Sputum culture is not part of the CDC criteria defining a VAP because sputum is virtually never culture negative in intubated patients, even without infection. You can use culture results of respiratory secretions to identify the Pathogen for your ICD10 Diagnosis (though positive blood or pleural fluid culture is considered more definitive).

Determining if an aspiration becomes secondarily infected

Here are some thoughts about how to figure out whether or not an aspiration event becomes secondarily infected:

The usual time course after the aspiration, of development of the lung inflammation and signs related to it, is typically hours to a day or more. And it is considered distinctly RARE for an actual infection due to an aspiration event to develop that quickly.

The usual time course of starting to see movement towards resolution of the signs of lung inflammation from noninfected aspiration pneumonitis, in the absence of secondary infection, is a few days. So, when a person who aspirates and is appropriately NOT given antibiotics starts to improve substantially within a few days, we take that as clinical confirmation that it wasn't infected.

On the other hand, the most clearcut kind of clinical evidence for secondary infection of a noninfectious aspiration pneumonitis is when the person starts to get better, and then after several days starts to get worse again.

The hardest situation in which to determine if a secondary infection has occurred, and whether to start antibiotics, is when the signs of lung inflammation (infiltrate, fever, leukocytosis) doesn't improves and remains bad or worsens.

No such thing as "presumed VAP"

Following CDC criteria, we will not code "presumed VAP".

No special rules for "CAP developing VAP"

There are no guidelines for CAP developing VAP, other than meeting the VAP criteria. For example, the idea of a CXR showing a "New and persistent" or "Progressive and Persistent" infiltrate allows for coding a second pneumonia on top of a first one.

VAP supersedes other pneumonia codes


If the patient qualifies for having a bacterial VAP, you code the VAP linked to the pathogen, and you do NOT have to code Pneumonia, bacterial.

  • Data collectors should follow criteria listed below regardless of what a physician writes in chart. If patient meets criteria VAP below, code as VAP. If patient does not meet all listed criteria, then do not code as VAP. It may qualify as a HAP or CAP.

Recent previous pneumonia

  • If a patient had any pneumonia previously during the same admission and then develops pneumonia again, meeting the VAP criteria, it is only a VAP if it is a new organism and has persistent or worsening infiltrates. If it is the same original organism, then the pneumonia has not completely been resolved, and you should NOT code it as a VAP.

Instructions regarding the attribution of a VAP

We report Ventilator Associated Pneumonia Rate based only on Acquired Diagnosis / Complication occurring in a critical care unit. Whenever one of those exists, we also collect and report QA Infection VAP. To make sure we have optimal data for this, the Data Processor and the Statistician have processes to make sure nothing is missed.

Attribution of the VAP to a Hospital Location

  • The infection is attributed to the location where the patient was on the date the infection became clinically evident -- EXCEPT if all elements of the infection are present within the first 48 hours of arrival, the infection is attributed to the location from which they were transferred.
    • An important consequence of this is that if on admission to the current unit it was NOT recognized that the patient has a VAP, but by these rules it is then figured out to be so --> THEN the correct coding of this VAP is as an Admit Diagnosis, not an Acquired Diagnosis.
  • The CDC case definition explicitly states that these rules should be followed -- that the physician’s statement of where the infection was acquired should not be substituted for these rules.
  • A new VAP that is identified first on a medicine ward must be attributed to the ICU that sent the patient.

See also: Iatrogenic, infection, central venous catheter-related bloodstream infection (CVC-BSI, CLI)#Attribution of a CLI to a Hospital Location

VAPs on medicine wards

  • The only way VAP can be coded on a medicine ward is if -- as per #1 - Ventilated 48 hrs -- the patient's Mechanical Ventilation (MV) ended that day or the prior day in an ICU and patient met all the criteria.
  • If the medicine collector identifies that a new pneumonia is present soon after arrival to the ward, they should let the ICU collector know to assess whether it was a VAP or not. VAP can be coded by the medicine collector as an Admit Diagnosis or Acquired Diagnosis following the usual rules, and should notify the sending ICU collector so they can ensure that the VAP is a Acquired Diagnosis for them and they have done the tmp information and notification process.

Long term ventilator patients with pneumonia

  • If a LTV patient is admitted from the community with a pneumonia, a Community Acquired Pneumonia (Community-acquired pneumonia (CAP) in ICD10) should be coded, not VAP, even though it is technically a VAP.
    • The rationale for this is that we only are wanting to tracking Hospital Acquired VAP's, not patients who have acquired an pneumonia while on long term home ventilators (LTV) in the community.
  • Of course, if the patient entered the hospital without a pneumonia, and develops one (and meets the VAP criteria), then that person would have a VAP we do want to code.

Reporting of complication when patients move units

The Ventilator Associated Pneumonia Rate we report is based only on Acquired Diagnosis / Complication occurring in a unit. If VAP is coded as an Admit Diagnosis, we check if the patient came from one of the ICUs where we collect data, and if so, make sure that the VAP is coded as a Acquired Diagnosis / Complication and QA Infection VAP there.

If a VAP Admit Diagnosis doesn't have a corresponding Acquired Diagnosis / Complication in the previous unit, the data processor will ask the collector to audit.

Iatrogenic Guideline

Iatrogenic injuries

Iatrogenic Infection

Regarding Attribution and Identification of Surgical Wound Infections

  • Note that these iatrogenic infections are attributed to the perioperative care for 30 days --- and for ONE WHOLE YEAR if related to an implanted device left in place
  • Our reference for this is this (specifically pages 9-10 to 9-14), and describes 4 entities:
    • ORGAN/SPACE SURGICAL SITE INFECTION -- without an implanted device left in place
    • ORGAN/SPACE SURGICAL SITE INFECTION -- with an implanted device left in place
  • For your purposes of whether such an infection is considered a Admit Diagnosis versus Acquired Diagnosis, use the timing rules as above
    • Here is an unusual consequence of this rule for surgical wound infections: Patient has a hip prosthesis put in 8 months ago. Admitted 1 month ago with pneumonia, and today is recognized to have an infection of that hip prosthesis. Despite the fact that the hip infection "seems" to have occurred well after this hospital admission, by the CDC rule it is actually a ORGAN/SPACE SURGICAL SITE INFECTION, and therefore it is attributed to the surgery one year ago, and so you should code it as a Admit Diagnosis even though the recognition of it was delayed for a whole month while in hospital.

Alternate ICD10s to consider coding instead or in addition

Pneumonia codes:

Candidate Combined ICD10 codes


Infections in ICD10 have combined coding requirements for some of their pathogens. Any that have antibiotic resistances would store those as Combined ICD10 codes as well. If the infection is acquired in the hospital, see Nosocomial infection, NOS. See Lab and culture reports for confirmation and details about tests. See Infections in ICD10 for more general info.

Possible Simultaneous Presence of Multiple Different Types of Infection in a Single Site

  • This refers to the situation where there may be simultaneous infection with multiple types of organisms -- e.g. 2 of bacteria, virus, fungus. While a classic example is a proven viral pneumonia (e.g. influenza) with a suspected/possible bacterial pneumonia superimposed, this kind of thing can occur in places other than the lungs, e.g. meningitis.
    • The "signature" of this is typically the patient being treated simultaneously with antimicrobial agents for multiple types of organisms. BUT don't confuse this with there being infections at DIFFERENT body sites.
  • As per our usual practice, we will consider a diagnosis as present if the clinical team thinks it's present and are treating it, with the exception that the team initially treated for the possible 2nd type of infection but then decided it likely was NOT present and stopped those agents.
  • And remember that Infectious organism, unknown is used when the the specific organism is unknown (this could be not knowing the TYPE of organism, or suspecting the type but not having identified the specific organism of that type), while when the organism has been identified but it's not in our bug list, THEN use Bacteria, NOS, Virus, NOS or Fungus or yeast, NOS.

Attribution of infections

See Attribution of infections

Related CCI Codes

Data Integrity Checks (automatic list)

Query check ICD10 Inf Infection req Pathogen must have oneCCMDB.accdbimplemented
Query Check Inf Pathogens must have Infection requiring pathogen or Potential InfectionCCMDB.accdbimplemented
Query s tmp QAInf tmp no dxCCMDB.accdbretired
Query check VAP admit must be from ICUCCMDB.accdbretired
Check VAP acquired only first encounterCentralized data front end.accdbdeclined
Query NDC VAP no TISSCentralized data front end.accdbdeclined
Query NDC VAP unacceptable dateCentralized data front end.accdbretired
Query NDC VAP AcqDX but NoVAP DateinTMPV2Centralized data front end.accdbretired
Query NDC VAP No AcqDX but VAP DateinTMPV2Centralized data front end.accdbretired

Data use

Used in:

Criteria change

The update in criteria in 2017-11 will likely lead to a higher rate of identifying VAPs. This needs to be noted in reports.

Related articles

Related articles:

Show all ICD10 Subcategories

ICD10 Categories: ANCA-associated Vasculitis (AAV), Abdominal trauma, Abortion, Acute intoxication, Addiction, Adrenal Insufficiency, Adverse effect, Alcohol related, Allergy, Anemia, Anesthetic related, Aneurysm, Antibiotic resistance, Antidepressant related, Aortic Aneurysm, Arrhythmia, Arterial thromboembolism, Asthma, Atherosclerosis, Awaiting/delayed transfer, Bacteria, Benign neoplasm, Breast disease, Burn, COVID, Cannabis related, Cardiac septum problem, Cardiovascular, Cerebral Hemorrhage/Stroke, Chemical burn, Chronic kidney disease, Cirrhosis, Cocaine related, Decubitus ulcer, Delirium, Dementia, Diabetes, Diagnosis implying death, Double duty pathogen, ENT, Encephalitis, Encephalopathy, Endocrine disorder, Endocrine neoplasm, Exposure, Eye, Female genital neoplasm, Fistula, Fracture, Fungus, GI ulcer, Gastroenteritis, Gastrointestinal, Gastrointestinal neoplasm, Hallucinogen related, Has one, Head trauma, Head trauma (old), Healthcare contact, Heart valve disease, Heme/immunology, Heme/immunology neoplasm, Hemophilia, Hemorrhage, Hepatitis, Hereditary/congenital, Hernia, Hypertension, Hypotension, Iatrogenic, Iatrogenic infection, Iatrogenic mechanism, Imaging, Infection requiring pathogen, Infection with implied pathogen, Infectious disease, Inflammatory Bowel Disease, Influenza, Inhalation, Intra-abdominal infection, Ischemia, Ischemic gut, Ischemic heart disease, Joint/ligament trauma, Leukemia, Liver disease, Liver failure, Lower limb trauma, Lower respiratory tract infection, Lymphoma, Male genital neoplasm, Mechanism, Meningitis, Metabolic/nutrition, Metastasis, Misc, Muscle problem, Muscles/tendon trauma, Musculoskeletal/soft tissue, Musculoskeletal/soft tissue neoplasm... further results


Use to be coded in TMP as QA Infection VAP. This was stopped 2019-Dec-31.