Initial antimicrobial treatment of hospital acquired pneumonia in adults: A conference report

Nosocomial or hospital acquired pneumonia is a serious and potentially fatal illness. It is generally defined as pneumonia that develops 48 h or more after admission to hospital; of all hospital acquired infections, it is the second most common but is responsible for the highest mortality rates from nosocomial infection (1,2). In a Canadian study, the overall mortality rate of hospital acquired pneumonia was 32%; 28% for nonintensive care unit patients and 39% for those in an intensive care unit (2).

Since nosocomial pneumonia is not a reportable disease in Canada, it is difficult to obtain accurate statistics on its true incidence. Figures are quite variable and range from five to 50 cases per 1000 admissions (3). Overall, nosocomial pneumonia rates for a Canadian tertiary care hospital are 9.5 per 1000 admissions. However, these figures vary greatly depending upon the individual service. Selected incidence figures per 1000 discharges are: gynecology, 2.5; general surgery, 5.8; surgical intensive care, 48.0; and medical intensive care, 67.0 (2).

Data on selected subgroups of patients with nosocomial pneumonia show that those who are bacteremic have the highest mortality rates, ranging from 48 to 58% (4) while mortality for infections caused by Pseudomonas aeruginosa may reach 70% (5). Studies of attributable mortality confirm the negative impact of nosocomial pneumonia on patient survival (6,7).

DEVELOPMENT OF GUIDELINES

Despite the seriousness of nosocomial pneumonia and its attendant sequelae, treatment is at times inadequate. This is usually due to a lack of appreciation of the likely pathogens, and the influence of risk factors in the individual patient. To develop a standardized approach to the initial antimicrobial treatment of hospital acquired pneumonia in Canada, a meeting was held in Toronto, Ontario in December 1992. The participants included specialists in infectious diseases, microbiology, respiratory medicine, intensive care medicine and surgery. Physicians from Canada, the United States and the United Kingdom took part, and the guidelines presented in this paper are a representation of the opinions of these experts.

The purpose of these guidelines is to help the practising physician in the choice of initial antimicrobial management of patients with hospital acquired pneumonia. These guidelines are not meant to deal with all situations, nor are they meant to be definitive. Their express purpose is solely to provide a rational approach to the initial antimicrobial treatment of hospital acquired pneumonia, and they are based upon careful consideration of various etiological, epidemiological and pathogenetic factors. The overriding concern in all cases is for the well-being and safety of the patient.

For these guidelines to be useful and workable, it was felt that they must be kept as practical, simple and user-friendly as possible. Had every clinical scenario been addressed, the guidelines would have become too cumbersome and complex and would soon be discarded. Accordingly, issues of pneumonia in severely immunocompromised patients such as those with organ transplants, neutropenic patients or human immunodeficiency virus-infected patients have specifically not been addressed, nor is tuberculosis considered. In Canada, such patients are often managed by physicians with specific expertise in these areas.

APPROACH TO PNEUMONIA

Nosocomial pneumonia may be approached in a number of ways, each with its own particular advantages and disadvantages. In developing these guidelines, it was decided to focus on what were felt to be the variables most likely to have an impact upon the potential microbial pathogens as well as the patient’s course and prognosis in hospital. The approach was based upon the following two variables:

• The severity of illness upon clinical presentation;

• The presence or absence of risk factors for specific microbial pathogens.

There are very few published studies that specifically define severity of illness for nosocomial pneumonia. Therefore, it was decided to extrapolate from data on community acquired pneumonia, and the following have been selected as criteria for ‘severe pneumonia’. A number of these features have been identified in a study of risk and prognosis in patients with nosocomial pneumonia (8). Patients may be considered to be severely ill if they present with any of the following: respiratory failure (PaO₂ less than 60 mmHg on an FiO₂ over 35%, with the exception of patients with chronic obstructive airway disease, who may be hypoxemic without pneumonia), respiratory rate 30/min or greater, sepsis with evidence of end organ dysfunction (severe sepsis), extrapulmonary septic complications, cavitation or involvement of more than one lobe on chest radiograph (9–11). For practical purposes, it is best to consider any patient with pneumonia who is being mechanically ventilated as having a severe infection.

It is clear from the literature and from clinical experience in general that certain events or procedures involving patients may act as risk factors for development of nosocomial pneumonia (2,12–14). Some of these include previous antimicrobial therapy, surgery, hospitalization in an intensive care unit, and use of ventilatory support (2). Among specific subgroups of patients with nosocomial pneumonia, such as those having had prior surgery or patients receiving ventilatory support, additional risk factors have been identified. In the surgery group, such factors include longer preoperative stays in hospital and longer operative procedures, as well as thoraco-abdominal surgery (12). In ventilated patients, additional risk factors include elevation of gastric pH, mechanical ventilation for more than three days and more than one intubation during ventilation (13,14).

For these guidelines, it was felt that the presence or absence of certain risk factors was critical to selection of initial antimicrobial management. Certain risk factors which might predispose to nosocomial pneumonia associated with a particular pathogen(s) are presented in Table 1 (15,16).

TABLE 1.

Initial treatment options for hospital acquired pneumonia

Clinical presentation – mild to moderate; risk factors present

Risks	Organisms	Drugs
Gross aspiration	Core organisms*†     +   anaerobes	Cefazolin + gentamicin	} ± Clindamycin   or Metronidazole
Thoraco-abdominal surgery		Second generation cephalosporin
		Nonpseudomonas third generation cephalosporin
		Ciprofloxacin
		or
		Ticarcillin-clavulanic acid
Diabetes	Core organisms     +   Staphylococcus aureus‡	Cefazolin + gentamicin
Coma		Second generation cephalosporin
Head injury		Nonpseudomonas third generation cephalosporin Ciprofloxacin
Prolonged hospitalization   and/or Prior antibiotics   and/or ICU admission	Core organisms (consider possible resistant Gram-negative rods and Pseudomonas aeruginosa)	See Table 3
High dose corticosteroids	Core organisms     +   legionella	Cefazolin + gentamicin	} + Macrolide
		Second generation cephalosporin
		Nonpseudomonas third generation cephalosporin
		Ciprofloxacin
Combination of risks		See Table 3 (severely ill)

ICU Intensive care unit;

^*Core organisms are listed in Table 2;

^†If the likely pathogen is an Enterobacter species, a cephalosporin should not be used regardless of intravenous suspected results;

^‡If methicillin-resistant S aureus is prevalent in your institution, consider adding vancomycin

Certain patients may also require admission to the intensive care unit. While a detailed discussion of this is beyond the scope of this paper, the following may serve as a guide. The presence of one or more of the following should be considered as a criterion for admission to the intensive care unit.

• Arterial oxygen saturation less than 90% on an FiO₂ of 0.5 using ventimask. In some patients, this may be preceded by rapid chest radiographic and arterial blood gas deterioration;

• Requirement for acute mechanical ventilation;

• Requirement for frequent chest physiotherapy and nursing supervision;

• Hypotension requiring inotropic support.

Basing the approach upon severity of illness and presence or absence of selected risk factors allows a flexible yet broad approach to initial treatment of hospital acquired pneumonia. It should be understood that these guidelines are meant only as an aid to the physician, and in all cases the individual physician dealing with pneumonia in a particular patient is ultimately in the best position to determine ideal therapy. Knowledge of the individual patient and of institution-specific microbial epidemiology and resistance patterns should be used to modify these guidelines whenever necessary or appropriate.

DIAGNOSTIC ISSUES

The etiological pathogens responsible for nosocomial pneumonia differ substantially from those which cause community acquired pneumonia. Organisms such as Streptococcus pneumoniae, Mycoplasma pneumoniae and Chlamydia pneumoniae are frequent pathogens In community acquired pneumonia while aerobic Gram-negative rods and Staphylococcus aureus account for the majority of hospital acquired pneumonia cases (2,17). In patients who have impaired consciousness or other neurological conditions which might predispose to aspiration, anaerobes should also be considered as potential pathogens.

Depending on the method used to determine microbial etiology, results may vary from study to study. Using the protected specimen brush, sputum cultures or blood cultures, however, Gram-negative organisms account for 61 to 75% of pathogens and S aureus for 22 to 33% (2). of the Gram-negative pathogens, P aeruginosa is the single most common organism, particularly in ventilated patients (2). Specific organisms that may be associated with particular risk factors are listed in Table 1.

The diagnosis of nosocomial pneumonia requires the synthesis of information obtained from a careful history, physical examination and appropriate laboratory tests and/or procedures. The history should identify any comorbid conditions and risk factors. The physician must also be careful to consider any noninfectious causes of fever and pulmonary infiltrates that may confound the picture. If possible, a good quality expectorated or suctioned sputum sample for Gram’s stain and culture should be obtained. Invasive procedures such as bronchoscopy are not usually necessary for most cases during the initial evaluation. Blood should be drawn for determination of a complete blood count and differential, and two sets of blood cultures should be collected as well. Routine radiological examination should include a posteroanterior and lateral chest radiograph. In patients who are intubated, a more aggressive approach to diagnosis may be necessary. In such cases, where expertise exists, use of both protected specimen brush technique and bronchoalveolar lavage (bal) are options to be considered (18). Gram’s stain of centrifuged bal fluid may then be used as a guide to initial empirical antibiotic therapy (18).

THERAPEUTIC ISSUES

In selecting drugs to be used for the treatment of hospital acquired pneumonia, a number of issues have been considered. Clearly, the severity of illness and the presence or absence of certain risk factors have influenced what are thought to be the likely pathogens and the suggested initial treatment. Activity and pharmacokinetic properties of antimicrobial agents and any other factors which might be of importance for the treatment of hospital acquired pneumonia were also considered. Options for initial oral therapy of mild cases are included as well.

Two issues in particular should be mentioned. In treating severely ill patients, concern about infection with P aeruginosa assumed prominence. Since mortality associated with this infection is considerably higher than is seen with other bacterial pneumonias, it was felt that treatment should be particularly aggressive (19,20). Data suggest that improved efficacy can be realized with combination therapy as opposed to single agent treatment of this condition, particularly if bacteremia is present (21,22).

There is also some concern about the use of aminoglycosides alone for the treatment of hospital acquired pneumonia. Since these drugs penetrate poorly into infected airways and may be inactivated by the low pH of secretions found in infected airways, it was felt they should not be the sole agent relied upon for the treatment of Gram-negative bacillary lung infections (23,24).

Based upon the variables discussed above, the various treatment options are presented in Tables 1, 2 and 3. Wherever possible, we have used classes of drugs rather than individual agents (Appendix). However, if only one drug in a given class was available or considered suitable, then that specific drug name has been used. Tables 1 and 2 deal with patients with mild to moderate infections who have either no unusual risk factors (Table 2) or in whom risk factors for specific pathogens may be present (Table 1). Table 3 deals with patients with severe hospital acquired pneumonia.

TABLE 2.

Initial treatment options for hospital acquired pneumonia

Clinical presentation – mild to moderate; no unusual risk factors

	Drugs
Organisms*	Intravenous	Orally
Klebsiella species	Cefazolin + gentamicin	Amoxicillin-clavulanic acid
Enterobacter species†	Second generation cephalosporin	Second generation cephalosporin
Escherichia coli	Nonpseudomonas third generation cephalosporin	Trimethoprim-sulfamethoxazole
Proteus species		Fluoroquinolone
Serratia marcescens
Staphylococcus aureus

With beta-lactam allergy, use ciprofloxacin or trimethoprim-sulfamethoxazole;

^*These represent the core organisms;

^†If the likely pathogen is an Enterobacter species, a cephalosporin should not be used regardless of intravenous suspected results

TABLE 3.

Initial treatment options

Clinical presentation – severe

Organisms	Pseudomonas aeruginosa
	Klebsiella species
	Enterobacter species
	Escherichia coli
	Proteus species
	Serratia marcescens
	Staphylococcus aureus*
	Legionella pneumophila†
Drugs	Intravenous
	Broad spectrum beta-lactam or fluoroquinolone with activity against P aeruginosa
	Piperacillin	} + Aminoglycoside‡
	Ceftazidime
	Imipenern/cilastatin
	Ticarcillin/clavulanic acid
	Ciprofloxacin

^*If methicillin-resistant S aureus is prevalent in your institution, consider adding vancomycin;

^†May be a nosocomial pathogen. If so, add a macrolide;

^‡Other combinations may be considered, eg, ceftazidime/ciprofloxacin

In all cases, whenever additional information such as culture and susceptibility data that may affect treatment become available, the attending physician should amend or modify the initial regimen as necessary. Attention must also be paid to supportive treatment including management of fluid and electrolyte balance, oxygenation and management of any complications that might arise.

Appendix. Antibiotics referred to in text

Drug class	Individual agents
I. Beta-lactams
a) Penicillins	Penicillin G*† Phenoxymethyl penicillin*
b) Anti-pseudomonal penicillins	Piperacillin†
c) Penicillins + beta-lactamase inhibitors	Amoxicillin-clavulanic acid* Ticarcillin-clavulanic acid†
d) Cephalosporins
First generation	Cefazolin†
Second generation	Cefuroxime*† Cefamandol†
Nonpseudomonas third generation	Ceftriaxone† Cefotaxime†
Third generation with anti-pseudomonas activity	Ceftazidime†
e) Carbapenem	Imipenem/cilastatin†
II. Macrolides	Erythromycin*† Clarithromycin*
III. Lincosamides	Clindamycin*†
IV. Fluoroquinolones	Ciprofloxacin*† Ofloxacin*
V. Aminoglycosides	Gentamicin† Tobramycin† Amikacin† Netilmicin†
VI. Miscellaneous	Trimethoprim-sulfamethoxazole*† Vancomycin†

^*Orally;

^†Intravenous