Indian Journal of Cancer Home 

[Download PDF]
Year : 2010  |  Volume : 47  |  Issue : 4  |  Page : 430--436

Use of antimicrobial agents and granulocyte colony stimulating factors for febrile neutropenia in cancer patients in a tertiary care hospital in India

V Roy1, D Saxena1, M Agarwal1, AK Bahadur2, B Mishra3,  
1 Department of Pharmacology, Maulana Azad Medical College, New Delhi, India
2 Department of Microbiology, Maulana Azad Medical College, New Delhi, India
3 Department of Radiotherapy, Maulana Azad Medical College and Associated GB Pant Hospital, New Delhi, India

Correspondence Address:
V Roy
Department of Pharmacology, Maulana Azad Medical College, New Delhi


Background: Use of antimicrobials (AM) and granulocyte colony stimulating factors (G-CSF) affect the outcome and cost of treatment of febrile neutropenia (FN). There are no studies describing the AM utilization pattern or the use of G-CSF and cost incurred on them in cancer patients with FN from India. Materials and Methods: A study was conducted in a tertiary care, teaching hospital in New Delhi, India, with the objectives of describing the utilization pattern of AM and G-CSF in cancer patients with FN. The efficacy and costs of AM and G-CSF prescribed were also assessed. Results: A total of 211 patients with FN were enrolled in the study. A majority of 207 (98.1%) were in the low-risk category. The average number of AM used per patient was 2.45 ± 0.02 and the AM exposure density was 1.19. All patients were administered five different combinations of AM regimens and G-CSF, irrespective of the risk category. No difference in the time to defervesence or in the recovery of ANC counts were observed with the different AM regimens. The average drug cost per febrile neutropenia episode (FNE) was Rs 4694.45 ± 296.35 (113.95 ± 7.19$). G-CSF accounted for 76.14 - 97.58% of the total costs. Conclusion: Large variations in the pattern of AM prescribed with routine use of G-CSF, irrespective of the risk status, was observed. Guidelines for the rational and cost-effective use of AM and G-CSF in patients with FN needed to be prepared. This was especially important as treatment was given free of cost to all patients admitted in the government health facility.

How to cite this article:
Roy V, Saxena D, Agarwal M, Bahadur A K, Mishra B. Use of antimicrobial agents and granulocyte colony stimulating factors for febrile neutropenia in cancer patients in a tertiary care hospital in India.Indian J Cancer 2010;47:430-436

How to cite this URL:
Roy V, Saxena D, Agarwal M, Bahadur A K, Mishra B. Use of antimicrobial agents and granulocyte colony stimulating factors for febrile neutropenia in cancer patients in a tertiary care hospital in India. Indian J Cancer [serial online] 2010 [cited 2022 Jul 7 ];47:430-436
Available from:

Full Text


Infections are a leading cause of death in cancer patients with febrile neutropenia (FN). [1] Standard treatment guidelines have been developed to rationalize the use of antimicrobials (AM) and granulocyte colony stimulating factors (G-CSF) for this condition. [2] However, in spite of the availability of the standard treatment guidelines for FN, variations in AM prescribed for FN continue to exist, resulting in differences in time to clinical response, duration of hospital stay, patient survival, and cost of therapy. [3],[4] The indiscriminate use of broad spectrum AM agents, not based on standard treatment guidelines, may increase the problem of antimicrobial drug resistance and the wasteful expenditure incurred on drugs. [5],[6] These are major issues of concern in a developing country like India.

It is estimated that in India there are approximately 2 - 2.5 million cases of cancer with seven lakh new cases being detected each year. [7] Treatment in most government health facilities is given free of cost to patients. Thus, the choice of medicines for treatment by doctors would have a major impact on the costs incurred by the hospital for purchase of medicines. Irrational use of expensive medicines may divert hospital funds from the purchase of other essential medicines for the patients. The newer AM and G-CSF are expensive medicines. There are no studies describing the AM utilization pattern or the use of G-CSF in patients of FN from India, and the costs incurred on the same. Hence, we decided to undertake this pilot study in order to assess the patterns of drugs used, specifically AM and G-CSF, in cancer patients with FN, and to evaluate the efficacy and cost of the drugs prescribed for the treatment of FN, in a tertiary care teaching hospital in India.

 Materials and Methods

The study was designed as an observational, prospective, cohort study. It was conducted in the Clinical Pharmacology Unit and Department of Oncology and Radiotherapy of a local Hospital, in New Delhi, India, from January 2007 to December 2007.

The study was approved by the Institutional Human Ethics Committee. A written informed consent was taken from all the patients enrolled in the study.

Patients of either sex, of any age, suffering from cancer with a diagnosis of FN, were included in the study. FN was defined as a condition in which patients have a single oral temperature ≥ 101° Fahrenheit (38.3° C) or a temperature ≥ 100.4° Fahrenheit (38.0° C) for ≥ one hour with neutrophil counts < 500 cells/mm [3] or < 1000 cells/mm [3] with a predicted decrease to less than 500 cells/mm [3] in the next 24 hours. [2] All patients who had received corticosteroids prior to admission were excluded from the study.

All patients fulfilling the study criteria were admitted to the ward. A detailed history was taken with regard to the patients' (i) demographic characteristics, (ii) history of illness, and (iii) treatment history (anti-cancer drugs taken, cycle number, drugs used for treatment of FN). A detailed examination of the patients was conducted. The risk status of the patients as low-risk or high-risk was determined based on a Specific Scoring Index, as given by the Infectious Disease Society of America (IDSA), 2002 [Table 1]. [2]{Table 1}

The following baseline investigations, that is, hemoglobin, total leukocyte count (TLC), differential leukocyte count (DLC), absolute neutrophil count (ANC), bacterial blood culture, and AM sensitivity of the isolate, were done on admission in all the patients. The TLC, DLC and ANC were repeated on days 0, 3, 5, 7, 10, and 14 or till the time of the patient's discharge. Fungal blood culture was done if the fever did not subside after seven days of antimicrobial therapy. The estimation of hemoglobin, TLC, and DLC was done on a Standard Automated Analyzer (Sysmex K-1000). ANC was determined using a stained peripheral blood smear in a Neubaur's chamber.

Details of the drugs administered for FN, specifically AM and G-CSF were recorded. These included the name of the drug, dose, route of drug administration, frequency of administration, and duration of days of treatment. Any change in the treatment prescribed was also recorded. An AM exposure was defined as a period of continuous administration of a single drug with no interruption in the administration of that drug for more than 24 hours. A therapeutic course, in contrast consisted of the simultaneous administration of one or more AM drugs, whether or not of the same chemical group, but given for the same clinical purpose, with no interruption in administration for more than 24 hours. [3] Antimicrobial exposure density was defined as the total number of antimicrobial exposure densities / total number of FN days.

The patient was followed up during his entire stay in the hospital and vitals, that is, pulse, blood pressure, temperature (six hourly, the last temperature was taken at 9:00 pm) were recorded. The patient's time to defervesence (in days) and time to recovery of ANC to above 500 cells and 1000 cells/mm [3] were recorded. Only a paracetamol tablet 500 mg, was to be administered for fever if required.

Efficacy was estimated by comparing the (i) time to defervesence (in days) and (ii) time to recovery of ANC counts (in days) with the different AM regimens. The outcome of the treatment was stated as successful or failed. Successful was defined as patient recovery from an episode of FN without any modification in the treatment regimens. The only exception being a treatment change because of a different AM sensitivity report. A regimen was classified as failed if the regimen had to be modified by the addition of one or more AM drugs or the patient did not recover from FN.

The drug costs of the AM, G-CSF, and paracetamol prescribed were calculated. The cost of the drugs was as per the rates at which they were procured by the hospital. The expenditure incurred on devices for parenteral administration was not considered for comparison of cost of different treatment regimens for patients of FN. The various AM treatment regimes used were categorized as 1, 2, 3, 4, and 5 and compared for efficacy, outcome, and cost. The AM treatment regimens were prescribed by the treating doctors in the hospital.

Statistical analysis

The results are presented as Mean ± Standard Error of Mean. For analysis of categorical variables, a two tailed Fischer Exact test was used. For inter-group comparison of quantitative variables with a normal distribution, Student's t test was used. Variables that were not distributed normally were compared using Kruskel Wallis. The correlation between the time to defervesence and the number of paracetamol tablets used and the correlation between the doses of G-CSF used and the time to recovery of ANC was measured using the Karl Pearson Correlation Coefficient. The results were considered significant if P < 0.05. All the calculations were performed using SPSS version-10.0.


A total of 211 subjects with FN were enrolled in the study from January 2007 till December 2007. The incidence of FN in cancer patients admitted to the Radiotherapy ward was 13.3%. The demographic characteristics of the patients are shown in [Table 2]. {Table 2}

The total number of days of FN was 731. The duration of the FN episode (FNE) ranged from two to five days. The average number of FN days per patient was 3.46 ± 0.13 days. The number of patients with ANC < 500cells/mm 3 was 27 (12.7%) and with ANC between 500 and ≤ 1000 cells/mm 3 was 184 (87.3%). All the patients had a temperature greater than 101 0 Fahrenheit at the time of admission. Blood culture was positive in only three patients. In all the three blood samples Escherichia coli was isolated, which was sensitive to ciprofloxacin, ceftriaxone, gentamicin, amoxicillin-clavulanic acid, cefuroxime, and amoxicillin.

All the patients with FN were administered a combination of AM and G-CSF. They were administered different AM regimens utilizing eight AM. These were ciprofloxacin, ceftriaxone, cefuroxime, cefoperazone-clavulanic acid combination, amoxicillin-sulbactam combination, metronidazole, fluconazole, and acyclovir. These were prescribed in the following regimens:

Regime 1: Ciprofloxacin + Metronidazole = 110 patients

Regime 2: Ceftriaxone + Metronidazole = 80 patients

Regime 3: Cefoperazone-Sulbactam + Metronidazole = 11 patients

Regime 4: Cefuroxime + Metronidazole = Eight patients

Regime 5: Amoxicillin-Clavulanic acid + Metronidazole = Two patients

The total number of AM agents used was 517 in a total of 211 therapeutic courses. The total number of AM exposure days in the hospital was 877. The average number of AM used per patient was 2.45 ± 0.02 and the AM exposure density was 1.19. The duration of AM therapy during the hospital stay ranged from two to five days. The majority of patients, 205 (97.1%), were administered AM parenterally by the intravenous route.

Metronidazole was administered to all the patients and antifungal drug fluconazole (150 - 200 mg) was administered to 13 patients. Only one patient was administered an antiviral drug, acyclovir tablet 800 mg, five times a day, for herpetic rash.

A total of 15 (7.1 %) patients were administered the paracetamol tablet (500 mg) with the average use of paracetamol tablet per patient being 0.53.

All the patients (211) both in the high- and low-risk category were administered G-CSF (filgrastim) immediately on admission via the subcutaneous route. The ANC at the time of administering G-CSF ranged from 100 cells/mm 3 to 1000 cells/mm 3

The average ANC at the time of G-CSF administration was 665 cells/mm 3 . The number of G-CSF doses administered to individual patients ranged from 1 to 3 . The total number of G-CSF doses administered were 460. The average number of G-CSF doses per patient was 2.18 ± 0.06. The characteristics of patients administered G-CSF are shown in [Table 3]. {Table 3}

The average time to defervesence per FN patient was 2.43 ± 0.13 days. The time to defervesence was significantly more with regimen 4 and significantly less with regimen 5, in comparison to the others [Table 4]. No correlation was found between the time to defervesence and the use of paracetamol tablets, with the different AM regimes.{Table 4}

The average time to recovery of ANC > 1000 cells/mm 3 was 3.29 ± 0.07 days. The time was minimum with regimen 5 and maximum with regimen 4 [Table 4]. The difference in the time for the ANC to recover was found to be significantly more with regimen 2 in comparison to regimen 1 (P < 0.05). For the statistical differences to achieve clinical significance, a greater and similar number of observations had to be made in different groups.

A total of 210 patients recovered uneventfully from FN. The fever subsided and the ANC increased to above 1000 mm 3 . The total number of patients who expired were three. In two of these patients the ANC increased, while in one the ANC did not increase. Two patients had received AM regimen 2 and one patient was on AM regimen 4.

The total cost of treating the FN episodes was Rs. 9,90,531.00 (24,042 $). The total cost per FNE was Rs. 4694.45 ± 296.35 (113.94 ± 7.19$).The cost for treatment of FN, per episode, with different regimens is shown in [Table 5]. The average cost of drugs per FNE was least with regimen 1 and maximum with regimen  4.{Table 5}

On comparison of the five regimens the cost of treatment per FNE was found to be significant between 1 and 2 and 4 and 5. (P < 0.05). The percentage of total cost spent on antimicrobials ranged from 2.4% with regimen 1 to 23.86% with regimen 3.

The percentage of total cost spent on G-CSF ranged from 76.14% with regimen 3 to 97.58% with regimen  1.


Cancer is among the leading causes of morbidity and mortality in India. [8] The exact incidence of FN in India is not known. There are no national guidelines available at present for treatment of FN in patients on cancer chemotherapy. The IDSA guidelines are being followed by well-recognized cancer treatment centers in India. Categorizing of patients based on risk status is important, as the treatment strategies for both are different. [2] Hence, we have used the IDSA guidelines for categorizing the patients into low- and high-risk. The incidence of FN in cancer patients admitted in the Radiotherapy ward was 13.3%, which was similar to that reported in other centers. [9] The majority of FN events (67.2%) occurred following the first cycle of chemotherapy. Previous studies have reported the incidence of FN events following the first cycle of chemotherapy to range from 11 - 67%. [8],[9]

The number of AM courses per FNE, average number of AM agents used per FNE, and the AM exposure density in this study were lesser than those reported in a study conducted over 18 tertiary centers in Europe, America, and Australia in which the number of AM courses varied from 1.6 ± 0.6 to 2.5 ± 1.5. [3] The overall use of AM for the treatment of FN in cancer patients appears to be lesser in our hospital in comparison to other health centers. [3]

In this study, the oral route of drug administration was used in only 2.8% of the patients, although a majority of patients, 98%, were in the low-risk category, with an ANC > 500 cells/mm [3] in 87.3% of the patients. Oral route for AM therapy has been recommended for low-risk category patients, as many studies comparing oral and parenteral route for administration of AM in low-risk FN patients have found comparable and successful outcomes. [10],[11],[12]

The most commonly used anti-bacterial agent was ciprofloxacin followed by ceftriaxone, cefoperazone / sulbactam combination, cefuroxime, and the amoxicillin / clavulanate combination. The choice of an antibacterial selected should depend on the likely causative microorganism of the FN. Fluoroquinolones are well-tolerated AM, with fewer adverse effects than other agents, but the problem is their limited activity against Gram positive organisms. [2] The next largest group used was cephalosporins, in 99 patients (46.9%). Three cephalosporins, that is, cefuroxime (second generation), cefoperazone, and ceftriaxone (third generation) were used. The rational for this could be their prominent activity against Gram negative organisms. We found that bacterial blood culture and AM sensitivity of three patients showed Escherichia. Coli, which were susceptible to ceftriaxone and cefuroxime. Another consideration could have been the cost. A five-day treatment course with ciprofloxacin was the cheapest (Rs. 140) followed by amoxicillin-clavulanate (Rs. 170), ceftriaxone (Rs. 170); cefuroxime (Rs. 684), and the cefoperazone-sulbactam combination (Rs. 1500).

All the patients were administered metronidazole. Metronidazole was usually administered when anerobic infection was suspected. However, empirical use of metronidazole in all patients appears not to be justified, as only 5% of the septicemic episodes in general are caused by anerobic bacteria. [13]

Thirteen patients were administered the antifungal drug fluconazole orally with no apparent reason for its administration. Only two of these patients were high-risk patients. According to the IDSA guidelines antifungal drugs should be administered to a patient of FN if the fever does not come down after seven days of AM therapy or if there is a documented fungal infection. In all the patients the temperature came down in lesser than seven days and fungal culture was not done.

The AM were prescribed to all the patients empirically. Although blood culture was sent in all the cases of FN, it was positive in only three patients in whom Escherichia coli was isolated, which was sensitive to the antibiotics the patients had been prescribed, so no change in their empirical treatment was done. All but one patient recovered from FN with the AM prescribed.

One reason for this could be that only one set of blood culture for bacterial testing was sent. It is recommended that in neutropenic fever, blood cultures from at least two samples, from different sites, for both bacterial and fungal growth, be sent, in addition to other tests. [1] This was not done in this study. Second, the diagnosis of an infection in a granulocytopenic patient could be difficult. [1] Many guidelines and recommendations for the appropriate use of AM agents in patients with FN are present. [2] However, the variations in the AM prescribed indicate that these are not being followed. [2],[14],[15],[16]

G-CSF was administered to all the patients along with the antimicrobials, irrespective of their risk status. The routine use of hematopoietic growth factors as adjunct therapy for the treatment of uncomplicated fever and neutropenia in all febrile neutropenia is not recommended. [17],[18] As per the guidelines, the use should depend on the risk category of the patients, their age, severity of marked neutropenia with counts < 500 cells/mm 3 , and patients with prognostic factors that are predictive of a poor clinical outcome. [14],[18],[19],[20]

G-CSF is an expensive drug. Our analysis showed that 76 - 97.5% of expenditure was incurred on account of the use of G-CSF. Antimicrobials accounted for only 2.4 - 5.6% of the total drug costs with most regimens and 23.86% with one regimen. Thus, in view of the high costs of G-CSF, the routine use of G-CSF in all patients of FN, irrespective of risk status, cannot be considered appropriate. Especially in a public health facility where medicines are given free of cost to patients.

On comparison of the efficacy of different AM regimens in terms of time to defervesence, it was found to be maximum in regimen 4 (3.13 days) and minimum in regimen 5 (2 days). It was not related to the use of paracetamol tablets. In both regimens 4 and 5, paracetamol tablets were not administered to any patient. With respect to the comparison in time to recovery of ANC above 1000 cells/mm 3 , it was observed that it was maximum with regimen 4 and minimum with regimen 1. However, it was significantly more with regimen 2 in comparison to regimens 1 and 5. No statistical significance was found between regimen 4 and others, probably because the distribution of the number of patients in different regimens was not uniform and adjustments made for statistical calculations could decrease the significance of any difference observed. The time to defervesence and time to recovery of ANC > 1000 cells/mm 3 was maximum with the regimen containing cefuroxime. The cost of a five-day course treatment with cefuroxime (regimen 4) was the second highest and the average cost per FNE was also maximum with regimen 4. The one patient with treatment failure, in the study, had also been prescribed cefuroxime. The AM regimen comprising of cefuroxime and metronidazole appeared to have lesser efficacy, with high costs.

The limitation of this study was that the bacteriological confirmation of infection was present only in three patients. However, the comparison between the different AM regimens did throw light on the existing patterns of treatment of FN in our center.

There appears to be a wide variation in the use of AM for treatment of FN in cancer patients, in our hospital. A combination AM therapy by the parenteral route, along with the routine use of G-CSF in all patients of FN, due to cancer chemotherapy, appears to be the practice. The selection of AM agents does not appear to be based on the hospital antibiotic sensitivity pattern. Excess expenditure is being incurred on routine use of expensive and inappropriate drugs. As the cost per FNE is very high, rational use of AM and G-CSF, with identification of low-risk patients, who can be treated as outpatients, with orally administered AM, would help in the better utilization of economic resources that could then be used for treating a greater number of patients.

There is an urgent need for guidelines to develop the appropriate use of AM agents and G-CSF for the treatment of FN in cancer patients in our hospital, along with continuing medical education on the same topic for the healthcare providers.


1Bergmann T, Sculler JP. Myelosuppression and infective complications. In: RL Souhami, I Tannock, P Hohenberger, JC Horiot, editors. Oxford textbook of Oncology. 2 nd ed. New York: Oxford University Press; 2002. p. 575-87.
2Hughes WT, Armstrong D, Bodey GP, Bow EJ, Brown AE, Calandra T, et al. 2002 guidelines for the use of antimicrobial agents in neutropenic patients with cancer. Clin Infect Dis 2002;34:730-51.
3Harbarth S, Viot M, Beeler I, Klastersky J, Szucs T; CEMIC study group. Variation in antimicrobial utilization for febrile neutropenia in cancer patients. Infection 2000;28:375-8.
4Elting LS, Rubenstein EB, Rolston K, Cantor SB, Martin CG, Kurtin D, et al. Time to clinical response: An outcome of antibiotic therapy of febrile neutropenia with implications for quality and cost of care. J Clin Oncol 2000;18:3699-706.
5Kunin CM, Tupasi T, Craig WA. Use of antibiotics: A brief exposition of the problem and some tentative solutions. Ann Int Med 1973;79:555-60.
6Howard DH, Scott RD 2 nd . The economic burden of drug resistance. Clin Infect Dis 2005;41:S283-6.
7K Park. Cancer. Park's Textbook of Preventive and Social Medicine, 19 th ed. Jabalpur India: M/S Banarsidas Bhanot; 2007. p. 318-27.
8Meza L, Baselga J, Holmes FA, Liang B, Breddy J. Incidence of febrile neutropenia (FN) is directly related to duration of severe neutropenia (DSN) after myelosuppressive chemotherapy. Proc Am Soc Clin Oncol 2002;21 Abstract 2840.
9Vogel CL, Wojtukiewicz MZ, Carroll RR, Tjulandin SA, Barajas-Figueroa LJ, Wiens BL, et al. First and subsequent cycle use of pegfilgrastim prevents febrile neutropenia in patients with breast cancer: A multicenter, double-blind, placebo-controlled phase III study. J Clin Oncol 2005;23:1178-84.
10Paul M, Soares-Weiser K, Leibovici L. Beta lactam monotherapy versus beta lactam-aminoglycoside combination therapy for fever with neutropenia: systematic review and meta-analysis. BMJ 2003;326:1111-5.
11Winfried VK, Alain C, Robrecht de B, Langenaeken J, Paesmans M, Gaya H, et al. Oral versus intravenous empirical antimicrobial therapy for fever in patients with granulocytropenia who are receiving cancer chemotherapy. N Engl J Med 1999;341:312-8.
12Hidalgo M, Hornedo J, Lumbreras C, Trigo JM, Colomer R, Perea S, et al. Outpatient therapy with oral ofloxacin for patients with low risk neutropenia and fever: a prospective, randomized clinical trial. Cancer 1999;85:213-9.
13Brown EA, Talbot GH, Provencher M, Cassileth P. Anaerobic bacteremia in patients with acute leukemia. Infect Control Hosp Epidemiol 1989;10:65-9.
14Segal BH, Baden LR, Brown AE, Casper C, Dubberke E, Freifeld AG, et al. Prevention and treatment of cancer related infections. J Natl Compr Canc Netw 2008;6:122-74.
15Cameron D. Management of chemotherapy-associated febrile neutropenia. Br J Cancer 2009;101:S18-22.
16Pascoe J, Steven N. Antibiotics for the prevention of febrile neutropenia. Curr Opin Hematol 2009;16:48-52.
17Ozer H, Armitage JO, Bennett CL, Crawford J, Demetri GD, Pizzo PA, et al. 2000 update of recommendations for the use of hematopoietic colony-stimulating factors: Evidence-based, clinical practice guidelines. Clin Oncol 2000;18:3558-85.
18Aapro MS, Cameron DA, Pettengell R, Bohlius J, Crawford J, Ellis M, et al. EORTC guidelines for the use of Granulocyte stimulating factor to reduce the incidence of chemotherapy induced febrile neutropenia in adult patients with lymphoma and solid tumors. Eur J Cancer 2006;42:2433-53.
19Hackshaw A, Sweetenham J, Knight A. Are prophylactic haematopoietic growth factors of value in the management of patients with aggressive non-Hodgkin's lymphoma? Br J Cancer 2004;90:302-5.
20Kelly S, Wheatley D. Prevention of febrile neutropenia: use of granulocyte colony-stimulating factors. Br J Cancer 2009;101:S6-10.