Special Feature
Fever and Neutropenia: More Challenges than Changes in the Last 20 Years
By J. Peter Donnelly, PhD
The number of neutropenic patients has in-creased considerably since Pizzo wrote his first editorial on fever in the immunocompromised host in 1983. Not only are there more patients being treated for malignancies but also more are receiving bone marrow transplants.1 These patients are compromised at virtually every level of their immunity—phagocytic, cellular, and humoral—and also have breaches to the integument affecting the skin and mucosal defense barriers. This gives rise to complications that may not be infectious but that are nonetheless accompanied by fever. But, because pyrexia is often the only sign of infectious disease, empirical antimicrobial therapy is still started promptly once an attempt at diagnosis has been made.
Risks Differ
All patients who are profoundly neutropenic (an absolute neutrophil count < 500 per cubic millimeter, < 0.5 ´ 109 L) and who develop a fever (temperature > 101°F or > 38°C sustained for several hours or a single temperature > 101°F or > 38.3°C) should be assumed to have a potentially life-threatening infection and treated accordingly until proven otherwise. This is still the most important principle governing management of potentially infectious complications that occur during neutropenia. At the same time, it is generally acknowledged that not all patients with neutropenia are at equal risk. For instance, those who develop neutropenia after cytotoxic chemotherapy or immediately after preparation for transplantation nearly always have concurrent breaches of the physical defense barriers, particularly the mucosa of the oral cavity and gastrointestinal tract. Such damage allows potentially pathogenic microorganisms that reside on these surfaces to establish local infection and gain entry into the tissues. Patients in this category should always be given empirical therapy promptly and in a hospital under the watchful eye of experienced doctors and nurses. By contrast, other less severely compromised individuals may require nothing more than careful observation and "watchful waiting" as outpatients. Several approaches to discriminating those at risk from those who are not at risk have been proposed as discussed by Rolston (see Table 1) but, as yet, none has been tested thoroughly enough nor for long enough to determine the true costs and principal benefits.2 Nonetheless, there is a clear desire to allow as many patients as possible to be managed at home rather than in hospitals.
| Table 1-Paradigm for Risk-Based Therapy of Febrile Neutropenic Patients | |||||||||
| Risk Group | Neutropenia | Malignancy | Procedure | Comorbidity | Clinical State | Anticipated Response to Empirical Therapy | Treatment Place | Treatment Approach | Treatment Administration |
| High risk | > 14 days | hematological | allogeneic HSCT, autologous HSCT, chemotherapy | substantial | unstable | slow | in-patient | empirical therapy until end of febrile episode | parenteral |
| Moderate risk | 7-14 days | solid tumor | autologous HSCT | minimal | stable | rapid | in-patient with early discharge | empirical therapy until end of febrile episode, then follow-up as outpatient | parenteral or parenteral then oral |
| Low risk | < 7 days | solid tumor | chemotherapy | none | stable | — | outpatient | appropriate out-patient treatment | oral, parenteral, then oral or parenteral |
| Source: Rolston KV. Clin Infect Dis 1999;29:515-521. | |||||||||
Pathogens
At first sight, the range of potential opportunistic and professional pathogens capable of causing infection in neutropenic patients is daunting, since virtually any organism can become invasive if host defenses are severely impaired. But, in practice, a few species of bacteria and fungi predominate. Bacteria can represent an immediate threat but there has been a remarkable and consistent change in the species of bacteria responsible for infection. Twenty years ago, the gram-negative bacilli such as Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa accounted for almost two- thirds of episodes of bacteremia and were rightly feared for the havoc they wreaked. Now the reverse is true, with the gram-positive bacteria predominating.3 (See Figure 1.) In fact, the skin commensal coagulase-negative staphylococci (mainly Staphylococcus epidermidis) and the oral viridans (or alpha-haemolytic) streptococci (principally S. mitis and S. oralis) are now the two most frequent isolates from blood cultures. The steady rise to prominence of the skin staphylococci can be attributed to the increased use of indwelling intravenous-access devices that remain in place for weeks and even months, whereas the occurrence of oral mucositis induced by more intensive chemotherapy seems to have allowed the viridans streptococci to make their debut. It must also be acknowledged that the shift away from the gram-negative bacilli toward these bacteria also coincided with more extensive use of co-trimoxazole (trimethoprim-sulfamethoxazole) and especially the fluoroquinolones such as ciprofloxacin for antibacterial prophylaxis. These drugs are effective in selectively decontaminating the gastrointestinal tract of gram-negative bacilli but they also exert considerable selective pressure on the commensal flora favoring the less susceptible bacteria such as those that reside on the skin and mucous membranes. It must also be said that these changes have yet to become apparent beyond the borders of northern Europe and North America, since in developing countries the old foe, gram-negative bacilli, still reigns supreme.
Initial Evaluation of Fever
The initial evaluation of a febrile, neutropenic patient should include a thorough inventory of the body to disclose sites of potential infection, particularly the oral cavity, lungs, gastrointestinal tract (including the perineal area), skin, and soft tissues. True, there is no clinical evidence whatsoever of such infection in up to two- thirds of cases, but the effort can be rewarding when a site is identified, such as an exit site infection of a central venous catheter or an anal fissure. These would be cases for considering including a focused-spectrum antibiotic in the empirical regimen.
Which Empirical Regimen?
There has been a protracted dispute about whether combination therapy is obligatory or monotherapy will suffice. In many respects the debate is more a matter of taste than substance since there are several antibiotics, such as the cephalosporins, ceftazidime and cefepime, and the carbapenems, imipenem and meropenem, that clearly are effective as single agents for empirical therapy (see Table 2).4 However, with the changing pattern of pathogens the issue deserves to be looked at afresh but there seems little willingness to do so. Instead, most physicians prefer to complement the initial regimen when they know what the cause of the fever is and some even when they do not. However, the fact remains that no matter which regimen is used, sooner or later other antimicrobial agents will be added to complement the core regimen. In many cases, the choice of drug will be driven by persistent fever that is as refractory to antibiotics as it is to diagnosis and is naturally assumed to be the result of a cryptic infection due to gram-positive cocci such as the coagulase-negative staphylococci. Thus, vancomycin will be chosen (or teicoplanin in Europe). Others will fear the worst and prescribe amphotericin B assuming that a fungus like Aspergillus fumigatus is lurking in the lung shadows. Many of these cases of persistent fever will remain unexplained and there will be no evidence to support the anxiety of the physician. But instinct and experience often provide better results at the bedside than guidelines arrived at by consensus. One issue does seem clear, however. Vancomycin (or teicoplanin) is seldom necessary upfront.4 These drugs contribute little in managing persistent fever, they can always be added when infection is clinically or microbiologically attributable to gram-positive bacteria and, importantly, the specter of vancomycin-resistant enterococci looms large and has already proved troublesome in some centers. So, the lesson here is to stay your hand and use vancomycin and teicoplanin sparingly.
| Table 2-Outcome of Monotherapy for the Initial Treatment of Fever During Neutropenia | |||||||||
| Percent of Episodes | |||||||||
| Success with monotherapy | Cefepime | Ceftazidime | Meropenem | Ceftazidime plus amikacin | |||||
| Overall | 53 | 55 | 56 | 52 | |||||
| MDI | 47 | 43 | 43 | 42 | |||||
| Glycopeptides added | 22 | 27 | 33 | 42 | |||||
| Glycopeptides added for MDI | 21 | 36 | - | - | |||||
| Further infections | 10 | 15 | 12 | 12 | |||||
| Source: Ramphal R. Clin Infect Dis 1999;29:508-514. | |||||||||
Resistant or Not?
Disturbingly, at least in North America, resistance to several antibiotics now exceeds 10% (see Figure 2), making it difficult to choose a made-to-measure, one-size-fits-all empirical regimen.6 The situation in Europe is more varied and resistance seems less pronounced. But the principal message to convey is that each institution should know its own resistance patterns and select empirical regimens as well as the most reliable agents for complementing this. Also, there should be a protocol to guide antibiotic use in each unit and a policy throughout the hospital. Regular antimicrobial surveillance of pathogens should be in place and the alarm should be raised whenever resistance rates start drifting in an upward direction. This, of course, requires resources that providers may be unwilling to commit. Indeed, it may seem cheaper just to choose the broadest regimen possible and apply it to every situation in the forlorn hope that overall costs will be contained and no one will suffer unduly. History suggests otherwise—blind, profligate use of antibiotics is dangerous, and has undoubtedly provided the greatest impetus for crafty bacteria to pull out all the stops and reinvent themselves in resistant clones. Moreover, it is no accident that high resistance rates run parallel with high consumption of antibiotics that only the wealthiest of countries can afford.
Beyond Managing Fever
Patients with prolonged immunosuppression may have several febrile episodes or persistent fever despite empirical therapy. One approach is to change therapy if there is no immediate response or by the next ward round, usually by increasing the number of drugs in direct proportion to the level of anxiety. Sometimes the regimen is repeatedly changed for as long as fever persists. This has been variously termed planned progressive therapy, and even escalatory therapy, and is often justified by the fact that with each change in therapy more patients become afebrile, supporting the belief that each change adds an incremental value to the ultimate response rate. However, in many cases, the putative increment is no more than what would be expected were no changes to therapy made.
Many patients with prolonged neutropenia require continued antimicrobial treatment but this does not exonerate the physician from carefully examining the patient with neutropenia, at least daily, for any signs and symptoms of progressive initial infection or new further infections which may develop over time. For instance, blood cultures should be done every 3-4 days for as long as fever persists. But it is remarkable how little attention is paid to the practice of microbiology in these articles. For instance, while we are still heavily reliant on blood cultures, we don’t take enough blood. Nor do we obtain proper specimens when we do see lesions on the skin or in the mouth and prefer to order a swab, having briefly entertained the idea of taking a biopsy only to dismiss it on the grounds that the results will prove of little value. True, we are now more inclined to ask for a bronchoscopy when a pulmonary infiltrate develops but remain in doubt as to its value, especially when negative, and so continue to treat empirically. In fact, although the empirical approach is based on the principle that fever is assumed infectious until there is proof beyond all reasonable doubt to the contrary, antimicrobial treatment will continue regardless of the results. The advent of better imaging techniques, such as high-resolution computed tomography and indirect means of detecting pathogens, should make it easier to diagnose infectious complications. For instance, microbiologically defined infections can now be diagnosed by detecting antigens such as Aspergillus galactomannan and the presence of microorganisms can de determined using PCR techniques. But whether they will bring about a change in attitude, only time will tell.
References
1. Pizzo PA. Fever in immunocompromised patients. N Engl J Med 1999;341:893-900.
2. Rolston KV. New trends in patient management: Risk-based therapy for febrile patients with neutropenia. Clin Infect Dis 1999;29:515-521.
3. Zinner SH. Changing epidemiology of infections in patients with neutropenia and cancer: Emphasis on gram-positive and resistant bacteria. Clin Infect Dis 1999;29:490-494.
4. Ramphal R. Is monotherapy for febrile neutropenia still a viable alternative? Clin Infect Dis 1999;29: 508-514.
5. Feld R. Vancomycin as part of initial empirical antibiotic therapy for febrile neutropenia in patients with cancer: Pros and cons. Clin Infect Dis 1999;29:503-507.
6. Jones RN. Contemporary antimicrobial susceptibility patterns of bacterial pathogens commonly associated with febrile patients with neutropenia. Clin Infect Dis 1999;29:495-502.
Which of the following is correct?
a. In developed countries, aerobic gram-negative bacilli remain a more common cause of bacteremia in neutropenic cancer patients than do gram-positive cocci.
b. All patients with febrile neutropenia must be given empiric parenteral broad spectrum antibiotics.
c. Solid tumor patients with chemotherapy-related neutropenia expected to be of less than seven days duration may be managed as outpatients.
d. Patients with hematological malignancies and chemotherapy-related neutropenia expected to last for more than 14 days may be managed as outpatients.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.