Guidelines for Community-Acquired Pneumonia: A Tale of 2 Countries

Pneumonia remains a serious medical problem, and its diagnosis and management is the source of considerable controversy. To date, 3 sets of guidelines have been published in Canada and the United States to help physicians deal with this disease in an organized manner [1–3]. Initially, there were enthusiastic and often heated discussions about the value of such guidelines, but now data support the value of regimens recommended in the guidelines and show reductions in cost, mortality, and length of hospital stay [4–6]. The guidelines have also pointed out the glaring gaps in our knowledge of the epidemiology, diagnosis, and treatment of pneumonia and have helped to direct research efforts in these areas. The latest contribution to the literature of community-acquired pneumonia (CAP) is the 2 guidelines from the United States and Canada [7, 8]. The former represents the efforts of the Infectious Diseases Society of America (IDSA) and the latter is a joint effort by the Canadian Infectious Diseases Society and the Canadian Thoracic Society. John Bartlett (Johns Hopkins University School of Medicine, Baltimore) and I were stereoisomers of each other in these endeavors; Dr. Bartlett chaired the American effort and served on the Canadian panel, while I cochaired the Canadian opus and served on the IDSA board. It was a pleasure and a privilege to work with Dr. Bartlett and with my colleagues from across Canada and the United States. Here I compare and contrast the 2 guidelines and put them into proper perspective for the reader. All of us involved in the development of such documents are well aware of the fact that most physicians in primary care practice, internal medical, or general surgery who “read” such guidelines usually limit themselves to a quick perusal of the key tables. Physicians who specialize in infectious diseases and pulmonary medicine, who have a greater interest in pneumonia, are likely to review these guidelines more thoroughly. However, we decided that, although not everyone was likely to read such works in their entirety, it is still important to make all the relevant information available for those wishing to critically review the literature and, even more important, to grade the evidence for the critical reader. Both documents are about equal in length and, although organized somewhat differently, they cover essentially the same material, with some exceptions. The obvious difference is the inclusion, by the IDSA, of a discussion of pneumonia in the context of bioterrorism. This is a timely and important topic, and it is covered in an efficient and expeditious manner. Whereas it is certainly important to consider local epidemiology and in vitro susceptibility patterns in the treatment of any infectious disease, particularly when using empirical therapy, pneumonia resulting from a bioterrorism attack may represent the one exception to this rule. Were such a device to be detonated in Toronto or Ottawa, it is highly unlikely that the pathogen used would have been obtained locally, and it was thought that the IDSA statement dealing with bioterrorism would apply equally well to Canada and the United States. For this article, the easiest way to compare and contrast the 2 sets of guidelines is to disregard the superficial differences, such as organizational headings and categories, and to divide the subject material into the 4 key areas and compare the essential points in each of these areas. The obvious topics of concern are etiology and epidemiology, the site-of-care decision, diagnosis, and treatment. The epidemiology of the disease is not notably different in the 2 countries, and both sets of guidelines provide a brief general description of this subject. The discussion of etiology was handled somewhat differently, with the IDSA guidelines providing a more general description of the various pathogens, possible risk factors, and potential influence of seasonal and geographic considerations and the Canadian guidelines focusing on the etiology according to where the patient was to be treated (i.e., ambulatory vs. hospitalized or nursing home patients). It is clear, however, that the take-home message is similar, with both groups believing that Streptococcus pneumoniae is the single most important pathogen. The atypical organisms are important not merely as causes of pneumonia: they can affect all age groups, and Mycoplasma pneumoniae and Chlamydia pneumoniae are capable of causing severe disease as well as mild to moderate illness. The issue of infection caused by aerobic gram-negative rods is touched on as well. The IDSA guidelines include these bacteria among the many possible etiologic agents and state that comparison of the relative frequencies of the various organisms as etiologic agents is hampered by the wide range of operating characteristics of the tests used to detect these pathogens in the published studies. The Canadian document states that, although aerobic gram-negative rods may be relatively uncommon causes of CAP overall, they are still very significant pathogens in patients ill enough to require care in an intensive care unit. These pathogens certainly cannot be ignored, and an analysis of data from 59 pooled studies suggested an average mortality rate of 33% when these pathogens were the etiologic agents [9]. The site-of-care decision is one of the most important decisions made in the management of a patient with CAP. It has medical, economic, and social implications associated with it, and, because access to health care continues to deteriorate in Canada and remains a problem in the United States, this decision will become even more important. It often is obvious whether a particular patient may be treated at home or requires admission to hospital; the problem lies with those patients for whom the correct decision is not so immediately apparent. When physicians are asked about such decisions, they often respond “I need to see the patient before I can tell.” Whereas this is certainly laudable, it still does not explain the criteria that are used to make this all-important decision. “Gut feelings,” hunches, or other such nonspecific criteria cannot be taught to medical students or residents nor can they be quantified and properly assessed and validated. Because the decision to admit has a direct effect on the intensity and type of laboratory testing and may also influence the type of drugs used and their mode of administration and cost, it was thought important to address this issue. Both groups have endorsed the Pneumonia Patient Outcomes Research Team clinical prediction rule of Fine et al. [10] for use in making this all-important site-of-care decision. Both groups, however, are also careful to point out the weaknesses of this prediction rule and the fact that it was not originally developed as a means of triaging patients with CAP. Other weaknesses are apparent as well. This rule is dependent on a large number of prognostic variables, it may at times oversimplify the interpretation of certain important indicators, and it fails to consider issues such as cognitive impairment that may be critical for any given patient. Its major strengths, however, are that it has been validated as a mortality prediction rule and it provides a rational, reliable, reproducible, and consistent basis for the decision to hospitalize a patient. No one is suggesting that this rule be used in all cases. As stated above, it is often obvious whether a patient requires admission to hospital. However, this rule may be used as a quality control tool and may also be used when hospital beds are in short supply. If there are 3 medical beds available in a particular hospital but 7 patients are in the emergency department awaiting admission, the rule could be used to argue on behalf of a particular patient. For example, if a patient with CAP fell into class V of the classification of Fine et al. [10], with a 29% risk of mortality, a cogent argument could made for access to 1 of the 3 available beds. Before considering the diagnosis of CAP, let us briefly examine the 2 main types of therapeutic approaches available to clinicians. These are directed therapy and empirical therapy. The former represents the use of an antibiotic aimed at a specific pathogen, and the latter is essentially an educated guess in which a broader-spectrum drug(s) is used in the hope that it will cover the possible etiologic agents. Directed therapy is clearly the more desirable, because it is associated with less polypharmacy and therefore lower costs, fewer adverse drug reactions, and less antibiotic selection pressure with the attendant risk of antimicrobial resistance. Why, then, do we use empirical therapy so often? The reason is the shortcomings of the diagnostic tests available to us. The main diagnostic modalities and techniques are history and physical examination, chest radiography, sputum gram stain and culture, blood culture, serology, antigen tests, and “other.” I will compare the 2 guidelines according to their assessments of and recommendations for these methods. Both guidelines ultimately provide recommendations for the extent and type of diagnostic assessment based on the severity of illness as reflected by the site of care chosen by the physician (outpatient vs. inpatient). In both cases, recommendations are limited in the sense that not a single study on which the recommendations are based was done with use of an appropriate reference standard, that is, autopsy or tissue biopsy. History and physical examination. Clinical assessment of the patient, consisting of a detailed history and physical examination, should be done because it may help to elicit potentially important risk factors or epidemiological clues, and it helps to establish a baseline as well. However, such an assessment is less than ideal for either making the clinical diagnosis of pneumonia or helping the physician to determine the etiologic agent. Chest radiography. Both groups agreed on the use of chest radiographs, recognizing that there may be instances in which it may not be feasible to obtain them. Ideally, they should be ordered for all patients (outpatients and inpatients) for whom a diagnosis of pneumonia is considered. It may help not only in ascertaining that pneumonia in fact exists but also in assessing severity and identifying predisposing causes, such as carcinoma or structural airway disease. Gram stain and culture of sputum and culture of blood. Both the IDSA and Canadian documents are remarkably consistent in their recommendations regarding gram stain and culture of sputum and culture of blood. The pros and cons of the sputum gram stain are discussed, and both groups acknowledge that its value is debated. Cultures of expectorated sputum samples are neither sensitive nor specific, and their value is open to question. Results of blood cultures, on the other hand, are thought to be of much greater value in that their specificity is very high, and it has been documented that obtaining blood samples for culture within 24 h of admission to hospital for CAP is associated with a statistically significant reduction in 30-day mortality [11]. Both guidelines recommend that, for patients sick enough to require admission to hospital, sputum be obtained for gram stain and culture and blood be obtained for culture as well. For patients managed outside the hospital, the IDSA suggests that gram stain of sputum is desired and culture is optional, whereas the Canadians do not recommend these for most patients. It is thought, however, that the gram stain may be helpful in regions where significant pneumococcal resistance is found because that might influence the choice of initial therapy. Neither group recommends that blood for culture be obtained from such patients. Serology and antigen tests. Serology is not recommended by either group as part of the routine workup of CAP patients, because the results would not be available in time to help with the initial treatment decision. At present, the 2 antigen tests of greatest interest are the membrane assay to detect pneumococcal antigen in urine and the urine test for Legionella pneumophila serogroup 1. The IDSA recommends the S. pneumoniae antigen test as a complement to sputum and blood culture. However, in their table dealing with diagnostic tests, they do not specify whether it should be done for outpatients, inpatients, or both. The Canadian group does not comment on this test because of a lack of experience with it and a paucity of published data. Both groups recommend the urine test for Legionella antigen as part of the workup of patients with severe CAP. Other. Invasive tests such as bronchoscopy or percutaneous needle aspiration are not to be used routinely. Both groups are in agreement that such techniques should be reserved for selected patients, such as those with a fulminant course or those failing to respond to a standard antimicrobial regimen. The sections dealing with treatment are organized somewhat differently, and they each emphasize different aspects of this issue. The IDSA statement, for example, provides a more detailed discussion of the various drugs available to the physician, whereas the Canadian statement reviews more extensively the evidence from therapeutic trials. At first glance, the approach to treatment appears to differ, but, on closer inspection, they prove to be very similar. The 3 obvious differences are that the Canadian guidelines deal specifically with nursing home pneumonia, they provide first and second choices instead of listing “generally preferred” drugs, and they appear to stratify outpatients into various groups whereas the IDSA seems to consider outpatients as a single entity. The first 2 differences are certainly readily apparent, but the third does not stand up to closer scrutiny. In table 13 of the IDSA statement, “outpatients” is listed as a single heading; however, selected considerations are provided in the table and are discussed in the body of the manuscript as well. In the Canadian statement, outpatients are stratified into those without modifying factors, for whom a macrolide may be used, and those with modifying factors, such as chronic obstructive lung disease, who have or have not received antibiotics or steroids within 3 months, and those with suspected macroaspiration. The fluoro-quino-lones are reserved for those in the former category. The US statement indicates that some prefer to use macrolides or doxycycline for patients <50 years old without comorbidity and fluoroquinolones for those ≥50 years old or with comorbidity. For patients requiring admission to hospital, both statements divide patients into those being treated on a medical ward and those receiving care in the intensive care unit. Both use the risk of infection with Pseudomonas aeruginosa as a means of further categorizing patients into subgroups. Both guidelines recommend a fluoroquinolone as monotherapy or a β-lactam plus a macrolide for the ward patients and a fluoroquinolone or macrolide in combination with a β-lactam for the intensive care unit patients in whom infection due to Pseudomonas is not an issue. If pseudomonal infection is an issue, an antipseudomonal β-lactam plus ciprofloxacin is a regimen recommended in both guidelines. These guidelines are far from perfect; they represent an attempt to deal with the available information in a comprehensive and organized manner. They are living documents that must be regularly updated and revised to deal with our expanding knowledge of the etiology and epidemiology of CAP, the resistance issues, and the availability of newer, rapid diagnostic techniques. Both sets of guidelines are exhaustive and thorough evidence-based reviews of the literature dealing with CAP. Both have organized the relevant information into coherent and manageable forms so that the reader can assess the evidence and make an informed decision. They have codified and structured our approach to the management of the patient with CAP and, by so doing, have provided us with a baseline against which future treatment options and interventions can be measured. Ultimately, the main purpose of these guidelines is to provide physicians with suggestions for the diagnosis and treatment of patients with CAP. It is fascinating that both groups have formulated recommendations that are so remarkably similar.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.