Managing the child with fever and neutropenia in an era of increasing microbial resistance☆☆☆
Article Outline
Abbreviations: AHS , α-Hemolytic streptococci, MIC , Minimum inhibitory concentration, MIC90, Minimum inhibitory concentration, 90%
See related article,p 20 .
There has been remarkable progress in the care of children with malignancy. Once fatal in nearly all children, acute lymphoblastic leukemia and lymphomas are now cured in more than 80% of all children. Solid tumors, such as Wilm's and rhabdomyosarcoma, are now cured in 70% to 90%. As a representation of one of the major revolutions in medical history, thousands of children and adults can now look back on their experience with cancer as an undoubtedly painful, but mercifully finite, period in their lives.
Yet, as every physician and family with such experience knows well, cancer therapy is not without toxicity. Most chemotherapeutic agents currently in use target rapidly dividing cells, cancerous or not. Consequently, areas of the body in which rapid cell division is a normal event are likely to be damaged in the course of cancer chemotherapy. Hair follicles, mucous membranes, the gastrointestinal tract, and bone marrow are affected to various degrees by nearly all agents used to combat malignancy. Although hair loss may be most bothersome to a young cancer patient, it is damage to the bone marrow that is most worrisome for the physician.
Septicemia associated with profound neutropenia is a major cause of death in children who receive treatment for cancer. Pathogens proliferate rapidly in the neutrophil-deficient bloodstream, resulting in potentially overwhelming infections. There may be few localizing signs in the absence of neutrophils that would normally migrate to the site of infection and initiate an inflammatory response. Consequently, minor infections may not be apparent until they become systemic. Even in the absence of neutrophils, however, proinflammatory cytokines such as interleukin-1, tumor necrosis factor, and interleukin-6 are released from several cell types in response to bacterial products, and fever is found almost universally in neutropenic children with serious infection.
Because fever is often the only indication of imminent septicemia, elevated temperature in a child with neutropenia is an indication for urgent attention.1 Evaluation of such a child includes a thorough physical examination with attention to sites of intravascular access, the respiratory tract, gastrointestinal tract, the perineum, and skin. Laboratory investigations include a complete blood cell count, culture of urine, and cultures of blood obtained from a peripheral vein and indwelling catheters. Comparison of quantitative colony counts from blood cultures from these 2 sites may distinguish catheter-associated infection from another source. Automated blood culture systems currently in use will detect most fungi and anaerobic bacteria. Nevertheless, in the setting of profound neutropenia and compromised mucosal surfaces, it is advisable to culture blood specifically for anaerobic bacteria and fungi. In certain instances, imaging studies such as chest radiograph, ultrasonography, computed tomography, or magnetic resonance imaging of the abdomen, sinuses, or other sites may be indicated, particularly if fever is prolonged and no apparent source is identified.
Empiric antibiotic therapy is guided by knowledge of the most likely causative organisms and their likely antibiotic susceptibility pattern. Many regimens are effective,2 most of which are comprised of either 1 or 2 antibiotics that rely predominantly on the activity of a β-lactam agent. Ceftazidime plus an aminoglycoside is one regimen in wide use. It is in this regard that the article in this issue of The Journal by Bruckner et al3 gives reason for pause and careful reconsideration of the empiric approach to the febrile neutropenic child.
Previously, gram-negative organisms accounted for the majority of cases of septicemia in neutropenic children. Recent studies demonstrate the ascension of gram-positive organisms.4 Among the most common isolates, coagulase-negative staphylococci and α-hemolytic streptococci (AHS) account for a large proportion of isolates. Coagulase-negative staphylococcal bacteremia usually arises from contaminated intravascular catheters. AHS, often referred to as viridans streptococci, reside predominantly in the oropharynx and gastrointestinal tract. Damage to mucosal surfaces that often occurs during chemotherapy provides a portal of entry for AHS. Despite the lack of classic virulence factors, these organisms are sometimes associated with fulminant infections.5 Streptococcus mitis , in particular, can cause irreversible shock and respiratory distress syndrome, frequently leading to death despite initiation of appropriate therapy.6
Bruckner et al3 report their findings of penicillin and cephalosporin resistance among AHS isolated from the blood of children with cancer. The retrospective review included all AHS isolates obtained at their institution over 3.5 years, ending in June 1999; 38 isolates were obtained from 33 children. Eighteen patients had hematologic malignancies, 14 had solid tumors, and 1 had recently undergone bone marrow transplantion. All patients had indwelling central venous catheters, and in those cases where mucosal surface integrity was documented, approximately one half had mucositis.
Susceptibility of AHS isolates to β-lactam antibiotics was determined by the E-test (AB Biodisk, Solna, Sweden), a method now commonly used in clinical microbiology laboratories to determine minimal inhibitory concentrations (MIC) by use of an antibiotic gradient. On the basis of pharmacodynamic data and clinical experience, the National Committee for Clinical Laboratory Standards developed guidelines that relate MIC of an antibiotic with the likelihood of clinical success. With the use of these guidelines, a physician can define an AHS isolate as susceptible, resistant, or as having “intermediate susceptibility” on the basis of its MIC for a particular antibiotic (notably, no such guidelines exist for ceftazidime against AHS).
Of the 38 AHS isolates examined in the current report, 21% were susceptible to penicillin, 42% were intermediate, and 37% demonstrated high-level resistance. A somewhat higher percentage of isolates was susceptible to the thirdgeneration cephalosporins ceftriaxone, cefepime, and ceftazidime (53%, 42%, and 53%, respectively). However, when the antibiotic concentration inhibiting at least 90% of isolates (MIC90) was determined, ceftazidime was the least active (MIC90 of 128 μg/mL, compared with 32 μg/mL, 8 μg/mL, and 16 μg/mL for penicillin, ceftriaxone, and cefepime, respectively). Knowledge of MIC90 results are useful in guiding empiric therapy, and the data presented here suggest that β-lactam agents, particularly ceftazidime, would fail to eradicate a significant percentage of AHS isolates and, thus, clinical failures would ensue.
Bruckner et al3 describe 2 adolescents with neutropenia with persistent fevers who had AHS bacteremia despite at least 4 days of treatment with ceftazidime and gentamicin. One child died of AHS meningitis, and the other initially had a relapse and was ultimately cured after a prolonged course of vancomycin.
In their study, factors correlating with resistance to β-lactam agents included recent use of systemic antibiotics, number of infectious episodes, and length of time from the initial diagnosis of cancer. These findings make intuitive sense, and are similar to those from another recent study on AHS resistance.7 Species type also correlated with β-lactam resistance, in that species found in normal oral flora (Streptococcus mitis , Streptococcus oralis , Streptococcus salivarius , and Streptococcus sanguis ) were more likely to be resistant than nonoral species.
The findings presented here have important implications for the management of children with fever and neutropenia. Resistance to β-lactam antibiotics among AHS, like that seen with Streptococcus pneumoniae , results from alterations in penicillin-binding proteins, rather than from production of a β-lactamase. Consequently, addition of a β-lactamase inhibitor, such as sulbactam or tazobactam, to a β-lactam agent would have no effect on AHS susceptibility. Vancomycin is commonly used to treat serious AHS infections when the isolate is known to be β-lactam resistant because no other antibiotic available currently has predictable activity against such organisms. Imipenem, clindamycin, and quinupristin-dalfopristin have been found variously to have better activity against AHS than penicillin or cephalosporins. Yet, resistance to each has been documented.5, 8, 9
The present study raises the important question as to whether vancomycin should be added to the empiric regimen for treatment of children with fever and neutropenia. Although some studies demonstrated a higher treatment failure rate among patients infected with AHS who did not receive empiric vancomycin compared with those who did, overall mortality rates were not different.10 These observations are considered in the context of emerging vancomycin resistance among some gram-positive organisms, particularly enterococci. It is presumed that increased use of vancomycin will result in higher rates of its resistance. A study by the European Organization for Research and Treatment of Cancer and the National Cancer Institute of Canada11 suggested that vancomycin need not be part of initial therapy for these patients.
The results presented by Bruckner et al3 and others will affect this discussion. It is evident that β-lactam resistance among AHS is not restricted to a single institution or region. Rates of resistance appear to be increasing, and recent reports generally demonstrate higher rates than those reports of only a few years ago (although these are difficult to compare because none are from the same site).3, 7, 9 These observations might initiate a trend toward increased empiric use of vancomycin. In counterbalance to this approach, it should be recalled that fewer than 20% patients with febrile neutropenia will have documented bacteremia12; of these, approximately 25% will be caused by AHS. With the resistance rates presented here, approximately 1 of 40 children with fever and neutropenia could be expected to have bacteremia as a result of β-lactam-resistant AHS. Unfortunately, some of these cases will be fulminant and fatal. However, initial therapy with vancomycin by no means guarantees of a cure. The 2 cases presented by Bruckner et al3 highlight the fact that blood cultures may remain positive for AHS, or infection may relapse despite vancomycin therapy. A study by Marron et al13 reported that, among 8 patients with serious AHS bacteremia who were treated empirically with vancomycin, 5 died.13
With these data in hand, guidelines for the empiric use of vancomycin that were published 5 years ago by the Infectious Disease Society of America remain valid.2 They suggest that only select patients receive vancomycin as part of their initial therapy. Indications include (1) clinically obvious and serious catheter-related infections, (2) substantial mucosal damage, (3) known colonization with β-lactamresistant Streptococcus pneumoniae or Staphylcoccus aureus , (4) growth of a gram-positive organism from the blood (pending identification and susceptibility testing), and (5) hypotension or other evidence of cardiovascular dysfunction. On the basis of the data presented by Bruckner et al,3 recent treatment with systemic antibiotics might be an additional consideration.
Immediate evaluation and rapid initiation of broad-spectrum antibiotics are often lifesaving for patients with febrile neutropenia. However, antimicrobial resistance is the inevitable and vexing trade-off. Studies such as the one by Bruckner et al3 allow clinicians to monitor the degree of resistance and tailor therapy accordingly.
Fortunately, such a reactive approach to the management of fever and neutropenia will not persist indefinitely. In the short term, clinical and basic research will lead to development of chemotherapy protocols with fewer side effects and antibiotics with improved activity against otherwise resistant organisms. In the long term, we can anticipate medications that target the aberrant biology of individual cancers rather than gross phenotypes, such as rapidly proliferating cells. Advances during the next several decades will likely make fever and neutropenia, like hair loss and mouth sores, forgotten aspects of cancer therapy.
References
- . Fever in immunocompromised patients. N Engl J Med. 1999;341:893–900
- 1997 guidelines for the use of antimicrobial agents in neutropenic patients with unexplained fever: Infectious Diseases Society of America. Clin Infect Dis. 1997;25:551–573
- . High incidence of penicillin resistance among α-hemolytic streptococci isolated from the blood of children with cancer. J Pediatr. 2002;140:20–26
- . The changing epidemiology of infections at cancer hospitals. Clin Infect Dis. 1993;17(2 Suppl):S322–S328
- . Molecular relationships and antimicrobial susceptibilities of viridans group streptococci isolated from blood of neutropenic cancer patients. J Clin Microbiol. 1999;37:1876–1880
- . Septicemia and shock syndrome due to viridans streptococci: a case-control study of predisposing factors. Clin Infect Dis. 1992;14:1201–1207
- . High rates of resistance to cephalosporins among viridans-group streptococci causing bacteraemia in neutropenic cancer patients. J Antimicrob Chemother. 2001;47:87–91
- . In vitro activities of eight macrolide antibiotics and RP-59500 (quinupristin-dalfopristin) against viridans group streptococci isolated from blood of neutropenic cancer patients. Antimicrob Agents Chemother. 1996;40:2117–2120
- Nationwide German multicenter study on the prevalence of antibiotic resistance in streptococcal blood isolates from neutropenic patients and comparative in vitro activities of quinupristin-dalfopristin and eight ether antimicrobials. J Clin Microbiol. 2001;39:1928–1931
- Vancomycin, ticarcillin, and amikacin compared with ticarcillin-clavulanate and amikacin in the empirical treatment of febrile, neutropenic children with cancer. N Engl J Med. 1988;319:1053–1058
- . Vancomycin added to empirical combination antibiotic therapy for fever in granulocytopenic cancer patients. J Infect Dis. 1991;163:951–958
- . Fever and neutropenia in children with malignant disease. Pediatr Hematol Oncol. 1996;13:503–510
- . Serious complications of bacteremia caused by viridans streptococci in neutropenic patients with cancer. Clin Infect Dis. 2000;31:1126–1130
☆ Reprint requests: David B. Haslam, MD, Department of Pediatrics, Box 8208, 6th Floor MPRB, 4938 Parkview Place, St Louis, MO 63110.
☆☆ J Pediatr 2002;140:5-7.
PII: S0022-3476(02)88362-7
doi:10.1067/mpd.2002.120266
© 2002 Mosby, Inc. All rights reserved.
Refers to article:
- High incidence of penicillin resistance among α-hemolytic streptococci isolated from the blood of children with cancer
