WT Hughes

Abstract

The population of immunocompromised children is increasing due to prolonged survival of patients with cancer, congenital immunodeficiency disorders, organ transplant recipients and the AIDS epidemic. Bacteria, fungi and viruses of low virulence comprise the etiology of most opportunistic infections encountered in these patients. Febrile episodes in most immunocompromised patients require immediate antimicrobial therapy, often initiated empirically before a specific diagnosis is possible. Especially demanding of broad-spectrum antibiotics are patients with profound neutropenia (< 500 cells per cu m). Systemic opportunistic fungal infections are now adding greater complexity to the management of these patients.

"Opportunism" refers to the practice of taking advantage of circumstances, especially with little regard for principles or consequences. An "opportunistic organism" denotes a microbe capable of causing disease only in a host whose resistance is lowered. Thus, "opportunistic infections" are limited to the immunocompromised host. However, infections of the immunocompromised host are not limited to opportunistic organisms but may also include disease caused by the more virulent organisms encountered in the healthy competent host. The greater frequency of infections due to low virulence organisms in the immunocompromised host is probably because of the ready and constant exposure to the nonvirulent, commensual flora of the environment in contrast to a much more sparse distribution of virulent organisms.

Whether or not a child becomes infected upon exposure to a microbial organism depends upon a critical balance between the child's resistance mechanism and the number and virulence of the organism. This relationship is expressed in the following formula:

For example, an organism of low virulence such as Staphylococcus epidermidis, if delivered in a high number to an individual with low resistance, may cause life-threatening disease. Under similar exposure a host with normal resistance, would not be infected. This formula provides a guide in practice. If one keeps the number of opportunistic organisms at a low density in the environment of the immunocompromised host (as with hand washing, personal hygiene, etc.) infection will be less likely to occur. Furthermore, if one attempts to maintain the integrity of host resistance mechanisms (Table I) infection rates can be minimized. Such approaches as avoiding unnecessary breaks in the skin by venipuncture, catheter entry sites, trauma, etc.; limiting the use of antibiotics that might destroy the important ecology of the microbial flora; and, minimizing periods and magnitude of neutropenia help reduce the prevalence of infection.

Children at Risk for Opportunistic Infections

Major categories of entities that predispose to opportunistic infections include: congenital immunodeficiency disorders, acquired immunodeficiency disorders including the acquired immunodeficiency syndrome (AIDS), cancer and cancer therapy, organ transplant recipients, and severe protein calorie malnutrition. Advances in medical practice have extended the lives of such patients leading to an ever increasing population of immunocompromised patients. Concomitantly, the prevalence of opportunistic infections has increased (Fig.). Usually rather profound defects of the immune system persisting over a prolonged period of time are required to render the patient highly susceptible to an opportunistic infection. An absolute neutrophil count of 1000 per cu mm, while well below the range of normal, does not create an obvious risk for infection, but a count of less than 100 per cu mm places the patient at very high risk for a life-threatening infection. Another indicator of risk is the CD₄ T lymphocyte count. While the CD₄ lymphocyte count is greater than 500 per cu mm the child or adult is at low risk for certain opportunistic infections such as Pneumocystis carinii pneumonitis, a count of 200 per cu mm or less heralds a high risk category. A few days of prednisone therapy for an asthmatic episode causes little risk to the host but prednisone given continuously for several weeks greatly increases the risk for life-threatening infection.

Fig Schematic indicating increase in population of immuno-compromised patients and concomitant increase in opportunistic infections.

Opportunistic Organisms

The organisms listed in Table II comprise the vast majority of microbes that infect the immunocompromised host. Most of these organisms may be found as components of the usual microbial flora, making it difficult under some circumstances to determine if an isolate in culture represents the cause of disease or merely colonization. This determination must be made by the physician and not the bacteriologist in the laboratory.

Clinical Manifestations

The clinical features of an opportunistic infection depend in great part on how capable the patient is to exert an immune response. The severely neutropenic patient is not able to evoke a full acute inflammatory response, regardless of what the infecting organism may be. So, with bacterial infection of the meninges, no inflammation occurs, which in turn does not illicit a stiff neck and pleocytosis of the spinal fluid. An infection in the lung may not be visible by chest radiograph due to lack of the density of inflammatory cells around invading organisms. A cutaneous infection by pyogenic bacteria may show little induration and tenderness and may exude a clear transudate because no neutrophils are available to make pus and inflammation.

Topographically, in neutropenic patients portals of entry tend to be on mucosal surfaces and skin around the oral cavity and perineal areas, where the normal microbial flora is dense with bacteria.

Fortunately, the patient's ability to generate a febrile response is rarely significantly diminished in any of the immunocompromised hosts, regardless of the underlying cause. Also, few of the underlying conditions of the immunocompromised host which are not of infectious origin, will cause fever. So, the occurrence of fever in such a patient strongly indicates an infectious disease and should always be considered of infectious etiology until proven otherwise.

Diagnostic Approaches

Two general categories of tests used in the diagnosis of infectious diseases are of little value in the severely immunocompromised host. These are serological tests for antibody and skin tests with antigens to determine delayed hypersensitivity. However, if antibody titer increases are found or skin test reactions are positive they may be considered valid. It is the absence of responses that precludes reasonable sensitivity. Humoral antibody to Toxoplasma gondii, Epstein Barr virus, varicella-zoster virus, cytomegalovirus are helpful indicators to know that the host at some time in the past has experienced such an infection, a factor for consideration in certain circumstances such as bone marrow transplantation.

Blood cultures for bacteria and fungi are of great importance. If central venous catheters are in place one should culture each lumen separately as well as a peripheral vein. Quantitative cultures are helpful, if available, to differentiate organisms derived from the catheter blood and a systemic bacteremia.

Cultures of the rectum and throat in the absence of visable lesions are of use only to show colonization with Pseudomonas aeruginosa and opportunistic fungi which may be of predictive value in neutropenic patients. Urinary tract infections can only be diagnosed by urine culture and colony count in neutropenic patients.

A chest radiograph should be done both to detect any evidence of pulmonary infection and to serve as a baseline for pulmonary complications that might develop later during the episode of infection.

Diagnostic imaging studies with ultrasound and computed tomography of the abdominal organs are especially helpful in the diagnosis of systemic fungal infection.

Biopsy of skin lesions as well as deep organ lesions often provides a definitive diagnosis, or may exclude a suspected etiology from further consideration.

Treatment

Because the course of infection in the immunocompromised host may be rapidly fatal, it is often necessary to take an empirical approach in the initial antimicrobial therapy. Well-established schemes have been prescribed for the febrile neutropenic patient. So, the absolute neutrophil count is used to determine the approach to initial treatment.

If the neutrophil count is less than 500 per cu mm, the patient should be given broad-spectrum antibiotics immediately by the intravenous route in maximal doses.^1-3 Any one of three antibiotic regimens may be used: (1) monotherapy with either ceftazidime or imipenem, (2) two-drug combination without vancomycin, such as an antipseudomonal penicillin (ticarcillin or piperacillin) plus an aminoglycoside (gentamicin, tobramycin or amikacin); or (3) vancomycin plus one or two drugs from (1) or (2). If cultures and other diagnostic tests identify a specific causative agent, treatment can be modified accordingly using drugs suggested in Table III and IV. The initial antibiotic regimen is continued for at least 3 days and at this time the condition is reassessed. If cultures are sterile and no evidence of progressive disease is evident the physician may either continue the same antibiotics by the intravenous route; or, change to an antibiotic administered by the oral route, such as cefixime. Recent studies show the overall outcome of the switch to oral cefixime is equal to continuation of the intravenous antibiotics.⁴ The main advantage of oral antibiotics is early discharge from the hospital. If the patient is still febrile at day 3 and vancomycin was not included in the initial antibiotic regimen consideration should be given to adding this drug to the treatment. If by day 5 to 7 of treatment, the patient is still febrile and neutropenic, amphotericin B should be added because the risk for the emergence of a systemic, deep organ, fungal infection is about 30% of such patients. If after about 2 weeks no evidence of deep fungal infection has evolved and no lesions are discernible by CT scan, the amphotericin B may be discontinued.

Febrile episodes in nonneutropenic patients with no localized evidence of infection can often be evaluated with an initial battery of diagnostic tests before deciding a course of therapy.

Prevention of Opportunistic Infections

In addition to adherence to appropriate hygienic principles and avoidance of exposure to individuals with contagious illnesses, some patients may benefit from chemoprophylaxis. The efficacy for prophylaxis with an antimicrobial drug has been demonstrated for patients at risk for Pneumocystis carinii pneumonia and those who have prolonged periods of severe neutropenia.^5,6 Trimethoprim-sulfamethoxazole daily or even 3 days a week is the prophylaxis of choice for P. carinii pneumonia. This drug combination given daily is also effective in the prevention of bacterial infection and febrile episodes in neutropenic patients. Quinolones, such as ciprofloxacin or ofloxacin, are effective in preventing febrile episodes in adults,⁷ but the lack of studies in children and the potential for adverse effects related to joints preclude the use of these drugs for prophylaxis in infants and children at this time and until definitive studies are available.

Current Major Problems and Controversies in 1996

Several circumstances are currently of concern either because of newly evolving problems or because adequate data are not available to determine the most appropriate management.

1. Emergence of antibiotic-resistant bacteria

Penicillin and cephalosporin-resistant pneumococci, methicillin-resistant Staphylococcus aureus and penicillin-resistant viridans streptococcus are now encountered as causes of bacteremia pneumonia and other types of infection.⁸ Vancomycin is necessary for treatment of these infections. Paradoxically, an equally troublesome problem is the emergence of vancomycin-resistant enterococci, due in part to the more extensive use of vancomycin. Current approaches include efforts to limit indiscriminate use of vancomycin and the use of teicoplanin or other antibiotics as substitutes for vancomycin.

2. The use of colony stimulating factors (G-CSF and GM-CSF)

G-CSF and GM-CSF effectively elicit a granulocyte response in most neutropenic patients. Exactly which pediatric patients should be given these molecules for treatment or prophylaxis is not clearly delineated. Reasonable guidelines have been published and provide details on usage. Generally cancer patients with neutropenia should not routinely receive G-CSF or GMCSF.⁹ Otherwise, in selected cases with progressive infection unresponsive to antibiotics, systemic fungal infection, pneumonia and/or hypotension the administration of G-CSF may be indicated.⁹

3. Use of oral antibiotics

The long-held dogma that antibiotic therapy for febrile neutropenic patients should be solely by the intravenous route has been challenged and new studies support the use of oral antibiotics as a part of the therapeutic regimen. We have recommended the use of oral cefixime in the section on treatment of infection in neutropenic patients. This approach can significantly reduce the days of hospitalization.¹⁰

4. Increase in opportunistic fungal infections

While systemic infections from Candida species and Aspergillus species are becoming more common, other opportunistic fungi once very rare are now increasing in prevalence. These include infections from Fusarium, Drechslera, Penicillium, Cunninghamella, Trichosporon species and others. Generally these are poorly responsive to available antifungal drugs.¹¹

In dealing with immunocompromised patients one must be continually alerted to the possibility that any microbe from any source of the environment may be a potential cause of serious disease. Because of the variable nature of man and microbe one must stay abreast of new scientific developments.

References

1. Hughes WT. Treatment of established bacterial and fungal infections in patients with hematologic malignancy. In Wiemik PH, Canellos GP, Dutcher JP. Neoplastic Diseases of Blood 3rd ed. Churchill Livingstone 1996;1027-40.

2. DePauw BE, Peresinki SC, Feld R, et al. Ceftazidime compared with pipercillin and tobramycin for the empiric treatment of fever in neutropenic patients with cancer. A multicenter randomized trial. Intercontinental Antimicrobial Study Group. Ann Intern Med 1994;120:834.

3. Ramphal R, Bolger M. Oblon DJ, et al. Vancomycin is not an essential component of the initial empiric treatment regimen for febrile neutropenic patients receiving ceftazidime: a randomized prospective study. Antimicrob Agents Chemother 1992;36:1062-7.

4. Shenep JL, Flynn PM, Hetherington SV, et al. Continued intravenous antibiotic therapy versus early switch to oral cefixime in neutropenic children with cancer and unexplained fever. 1994. Proc. Infectious Diseases Society of America, Orlando, FL.

5. Hughes WT, Kuhn S, Chaudhary S, et al. Successful chemoprophylaxis for Pneumocystis carinii pneumonitis. N Engl J Med 1977;297:1419-26.

6. Verhoef J. Prevention of infections in the neutropenic patient. Clin Infect Dis 1993;17(Suppl 2):5359-67.

7. Liang RH, Yung RW, Chan TK, et al. Ofloxacin versus cotrimoxazole for prevention of infection in neutropenic patients following cytotoxic chemotherapy. Antimicrob Agents Chemother 1990;34:215-8.

8. Bochud PY, Eggiman Ph, Calandra Th, Van Melle G, Saghafi L, Francioli P. Bacteremia due to viridans streptococcus in neutropenic patients with cancer: clinical spectrum and risk factors. Clin Infect Dis 1994;18:25-31.

9. American Society of Clinical Oncology. Recommendation for the use of hematopoietic colony-stimulating factors: evidence-based clinical practice guidelines. J Clin Oncol 1994;12:2471-508.

10.Law RC, Doyle JJ, Freeman MH, King SM, Richardson SE. Early discharge of pediatric febrile neutropenic cancer patients by substitution of oral for intravenous antibiotics. Pediatr Hematol Oncol 1994;11:417-21.

11. Sugar AM. Empiric treatment of fungal infections in the neutropenic host. Review of the literature and guidelines for use. Arch Intern Med 1990;150:2258-64.