By Author
  By Title
  By Keywords

December 1998, Volume 48, Issue 12

Original Article

The Spectrum of Bacterial Infections in Febrile Neutropenic Patients: Effect on Empiric Antibiotic Therapy

Ikram A. Burney  ( Departments of Medicine, The Aga Khan University Hospital, Karachi. )
Badar J. Farooqui  ( Departments of Pathology, The Aga Khan University Hospital, Karachi. )
Tariq Siddiqui  ( Departments of Medicine, The Aga Khan University Hospital, Karachi. )
Mohammed Khurshid  ( Departments of Pathology, The Aga Khan University Hospital, Karachi. )

Abstract

The aim of this retrospective analysis was to look at the spectrum of bacterial isolates and their resistance patterns to the commonly used antIbiotics in the settmg of febrile neutropenia. A total of 127 bacteria were isolated from patients with acute leukemias, lymphoproliferative disorders, aplastic anaemia and various solid tumours. Fifty-four percent organisms were gram negative; while the rest were gram positive. E. coli, pseudomónas aeruginosa, staphylococcus aureus, enterococcus and streptococci were the commonly isolated organisms. Forty-eight percent organisms were isolated from blood, 16% from urine, 13% from wounds and superficial abscesses and 11% from respiratory tract. E. coli exhibited a great degree of resistance to the commonly used antibiotics, such as pipericillin (70%), ofloxacin (50%) and aztreonam (50%). Pseudomonas and kiebsiella also showed varying degree of resistance against the antibiotics. Staphylococcus aureus and staphylococcus epidennidis were almost universally resistant to penicillin and showed a variable degree of resistance to other antibiotices too. Compared to the previous reports, the pattern of bacterial isolates and their resistance to antibiotics has changed over the past years. Aminolgy­cosides and third generation cephalosporins seem to be the choice of antibiotics for the upfront manage­ment of febrile neutropenic patients (JPMA 48:364,1998).

Introduction

Chemotherapy results in neutropenia in a great majority of patients. The patients with hematological and lymphoid malignancies are particularly susceptible1. Of the total number of patients who develop neutropenia, about 80% have fever2. A febrile neutropenic patient requires empiric antibiotic therapy, which should be instituted as early as possible3. Delay in initiation of treatment may result in septicaemic shock and is often associated with mortality4. The vast majority of neutropenic febrile patients (75-80%) do not have a clinical focus of infection at the time of presentation5. Hence, empiric antibiotic therapy is formulated on the knowledge of common bacterial isolates and their sensitivity patterns. Gram negative infections were thought to be the most common offending agents and usually a combinationof aminoglycosides and beta lactam antibiotics result in defervescence1,2. However, in the past few years, with increasing usage of indwelling catheters and effective treatment of gram negative infections, there has been a shift in the spectrum of bacterial isolates5,6. Gram positive organisms are now isolated with increasing frequency and gram negative organisms display more resistance to B-lactam and other antibiotics. We report the patterns of isolates as well as their resistance to antibiotics frnm a single institution in Pakistan.

Patients and Methods

The data was collected retrospectively from all the febrile patients seen under the oncology service with culture proven bacterial isolates over a period of 15 months. Of the 379 bacterial pathogens isolated in this period, 127 were identified when the absolute neutrophil count (ANC) of patients was less than 500/W. The patients either had hematological or non- hematological underlying malignancies oraplastic anemia, Cultures were obtained from blood, urine, sputum, wound/abscess or any other identifiable focus of infection. Culture and sensitivity were carried out according to the routine procedure in the clinical laboratories of our hospital. Briefly, 5 ml of blood was drawn into each of brain-heart infusion bmth and thioglycollate broth and incubated at 37°C for 24 hours. Gram negative bacteria were identified using API 20E strips and gram positive bacteria by the catalose test, coagulase test, sensitivity to optochin and bacitracin discs. Antibiotic sensitivity was performed on diagnostic sensitivity agar plates by Kirby-Bauer method7.

Results

Over a period of 15 months, 127 bacterial isolates were cultured from a total of 85 patients while they were neutropenic and febrile. The majority of patients had acute myeloid leukemia (23/85) or various lymphopmliferative disorders (25/85) including acute lymphoblastic leukemia. Seventeen had aplastic anaemia and 18/85 bad solid tumors. One patient each had multiple myeloma and agranulocytosis. Gram negative bacteria accounted for 69/127 (54.5%) of the isolates. E. coli was the single most common organism isolated, followed by pseudomonas aeruginosa and Klebsiella. Details are shown in Table 1.

The resistance patterns of the commonly isolated organisms were studied against frequently prescribed antibiotics. The antibiotics included aminoglycosides, B-lactam antibiotics, monolactams, third generation cephalosponns and the quinolones.

Table II shows the resistance patterns of E. coli, pseudomonas and Kiebsiella against various antibiotics. E. coli were resistant to pipencillin in 70% cases and 50% were resistant to ofloxacm, ameonam and cefixime. Twenty-three peicent of the pseudomonas species were resistant to pipericillin, whereas 15% were resistant to aztreonam and ofloxacin.
Gram positive organisms comprised 46% (58/127) of the total bacterial isolates. Staphylococcus aureus was the commonest gram positive organism isolated (15/58), followed by streptococcal species (Table III).

Staphylococcus epidermidis were isolated on 9 occasions. All but one patient with bacteremia secondary to staphylococcus epidermidis had an indwelling venous access catheter at the time of the febrile episode.

Table IV shows the resistance patterns of staphylococcus aureus and epidermidis. Virtually all the organisms were resistant to penicillin.

Discussion

Grain negative aerobic bacteria have been the predominant- isolated organsism from febrile neutropenic patients1,3. Over the past few years there has been a shift of microbial flora from grain neptive to gram positive organisms at many cancer centers5,6. This could be attributed to bettercontrol of gram negative infections and more frequent use of indwelling venous access devices6. An earlier study at our institution revealed gram negative organisms to account for 70% of all infections8. The vast majority were pseudomonas which comprised 31% of all the isolates. As a result, a combination of amikacin and pipericillin was used commonly to treat prolonged (more than seven days) neutropenic febrile episodes, whereas ofloxacin was frequently used to treat neutropema expected to last less than seven days9.
The current study demonstrates that gram positive organisms have increased in frequency at this institution. Coagulase negative staphylococcus was isolated with increasing frequency, particularly from patients who had indwelling venous access devices. Gram positive rods of Bacillus species were likewise isolated from fourpatients. The majority of whom had indwelling catheters. The vast majority of patients in this series (78%) had either lymphoproliferative ormyelopmliferative disorders or had aplastic anaemia and as such had prolonged neutropenia. Fungi (candida and aspergillus) were cultured on 18 occasions from the same patient group during the study period. However, they were excluded from the overall analysis, since the aim of this study was to investigate the bacterial infections in neutropenic patients which are important in the early empiric antibiotic management.
The demonstration of nearly 50% of the bacterial isolates to be gram positive organisms is in agreement with several recent reports. Petrilli et al10 have shown gram positive cocci to account for 52.6% of the infections. Similarly Coullioud et al6 showed 56% of 1147 pathogens to be gram positive bacteria. However, majority of their patients had indwelling venous lines at the time of febrile episodes. In our study, 19/28 organisms isolated from patients with indwelling various catheters were gram positive.
A previous study at this institution showed pseudomonas to be isolated on 31% occasions8, the current study shows pseudomonas to account for only 11% of the isolates. E. coli were the most frequently isolated organisms(16%). An important features of this study is the demonstration of increasing resistance of E. coli to various antibiotics. Of the 20 isolates, 10 were resistant to quinolones, one of the frequently used drugs for the treatment of such infections11, particularly for those who have low risk neutropenia12. Pipericillin had been used routinely as an integral part of the neutropenic regimen in the past. Seventy percent of the E. coli were resistant to this antibiotic as well, while 50% E. coli were also resistant to aztreonam, a monolactam frequently used as an alternative to the ‘first line’ arnikacin and pipericillin. E. coli showed 5% and 15% resistance to Aminoglycosides and third generation cephalosponns respectively. Besides E. coli, resistance to quinolones, beta lactam, monolactams and third generation cephalosporin has also emerged forpseudomonas and klebsiella.
Another feature of this study is the fact that isolates obtained from any culture-directed site were included for overall analysis. Many of the studies have used only bacteremic episodes to look at the pattern of isolated organisms6’7. If bacteremic episodes alone were to be considered, 30/61 i.e., nearly half the bacteria were gram positive in this study. This is again consistent with the studies mentioned above. Although the majority of organisms were isolated from blood (48%), followed by urine (16%), wound/abscesses (13%) or sputum and tracheal aspirate (11%), occasionally micro-organism were also isolated from sites such as the ascitic fluid, pleural fluid, CSF etc. Hence, it is imperative that every effort should be made to obtain cultures from all possible sites, particularly as the bacterial spectrum has changed and new resistance patterns have emerged. Demonstration of antibiotic resistance resulted in frequent modifications of treatment in these patients.
It is concluded that the gram positive bacteria have increased as a cause of infection in neutropenic febrile patients, particularly in those with indwelling venous access devices. Multi-drug resistant strains of gram negative organisms have emerged rapidly. This may pose a great problem in early management of these patients. A combination of aminoglycosides and third generation cephalosporin seems to be appropriate for the upfront management at this time. Studies of this nature should be frequently repeated to document the changing pattern of bacterial isolates and the antibiotic resistance which has marked implications in the choice of early empiric antibiotic therapy.

References

1. Pizzo PA, Robichaud KJ, Wasley R, et al. Fever in the pediatric and young adult patient with cancer: A prospective study of 1001 episodes. Medicine, 1982,61:153-65.
2. Pizzo PA, Management of fever in patients with cancer and treatment induced neutropenia. N. Engl. J. Med. 1993;398.1323.31.
3. Pizzo PA, Robichaud KJ. Gill FA. et al. Empiric antibiotic and antifungal therapy for cancer patients with prolonged fever and granulocytopenia. Am. J. Med., 1982;72 101-11.
4. Bodey GP, Bukley M, Sathe YS, et al. Quantitative relationships between circulating leukocytesand infection inpatientswithacute leukemia. Ann. intern. Med., 1966;65:328-40.
5. Pizzo PA, Ladisch S, Simon RM. et at. Increasing incidence of Gram-positive species in cancer patients. Med. Pediatr. Oncol., 1978;5:241-44.
6. Coultioued D, Van-der AP, Viot M. et a!. Prospective multicentric study of the etiology of 1051 bactcremia episodea in 782 cancer patients. CEMIC French-Belgium study club of infectious diseases in cancer. Support Care Cancer, 1993;1 :34-46.
7. Bauer AW, Kirby WMM. Sheris JC, et a!. Antibiotic susceptibility testing by a standardized single diac method. Am. J. Clin. Pathol,1966;45:493-9.
8. Karim M, Khan W, Farooqui B, et at. Bactrial isolates in neutropenic febrile patients. J. Pak. Med. Assoc., 1991 ;41 :35-7.
9. Malik IA, Abbas Z, Karim M. Randornised comparison of oral ofloxacin alone with combination of parcnteral antibiotics in neutropenic febrile patients. Lancet, 1992;339: 1092-6.
10. Petrilli AS, Bianchi A, Kusano E, et at. Fever arid granulocytopenia in children with cancer: A study of 299 episodes with two treatment protocols in Brazil. Med. Pediatr. Oncot., 1993;2 1:356-61.
11. Malik IA, Khan WA, Karim M, et at. Feasibility of outpatient management of fever in cancer patients with low-risk neutropenia Results of a prospective randomized trial. Am. J. Med., 1995;98:224-30.
12. Tatcott JA, Siegel RD. Finberg R, et at. Risk assessment in cancer patients with fever and neutropenia: A prospective,two- center validation of a prediction rule. J Clin.Oncol 1992;10:316-22.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: