Early discharge from hospital after consolidation chemotherapy in acute myeloid leukemia in remission: febrile neutropenic episodes and their outcome in a resource poor setting

Treatment of acute myeloid leukemia (AML) involves administration of myelosuppressive chemotherapy administered after admittance to hospital.1 Admission to intensive nursing care units till bone marrow recovery leads to prolonged hospital stay. Quality of life and health care issues have made many cancer centers change to outpatient care even during high-risk phases of disease.2–6

In India most patients belong to poor socioeconomic backgrounds. There is an acute shortage of hospital beds. Early discharge after myelosuppressive therapy would promote better use of hospital resources, but the safety of this approach in these patients has not been established. We present our experience of the feasibility and safety of early discharge of patients with acute myeloid leukemia after consolidation chemotherapy.

All patients were induced with standard ‘3+7’ chemotherapy using a peripherally inserted central venous (PICC) line. After documentation of complete remission (CR) consolidation chemotherapy with 3 cycles of high dose cytarabine was given.

Eighty-three consecutive episodes of neutropenia after consolidation chemotherapy in 28 acute myeloid leukemia patients in remission were included in the study. Patients were divided into 2 groups.

Group 1

Outpatients

These consisted of patients discharged after the chemotherapy was completed to their own homes or temporary residential facilities, which did not have any medical, or home visit facilities. Criteria for inclusion were: (a) no fever or infection; (b) location of residence nearby; (c) ability to come to hospital within one hour if fever developed or condition deteriorated. They had telephone access to the study team.

Group 2

Inpatients

These were patients who remained in hospital after high dose Ara C (HiDAC) chemotherapy till recovery of neutrophil counts. The criteria of inclusion were: (a) inability to move to a residential place as specified under group 1 or (b) severe infective course during earlier chemotherapy.

All patients were given the following supportive therapy: prophylactic ciprofloxacin 500 mg twice daily and fluconazole 200 mg/day and simple instructions concerning hygiene. Blood counts were monitored twice a week for Outpatients and on alternate days for Inpatients. Outpatients were seen at least once a week in the outpatient department (OPD). Blood and platelet transfusion support was given in the day care center. Patients who developed fever were administered granulocyte colony stimulating factor (G-CSF). Fever was considered present if the temperature measured orally was ≥38° C on two occasions at least four hours apart during a 24-hour period or was ≥38.5°C on a single occasion. Neutropenia was defined as an absolute neutrophil count (ANC) of <0.5×10/L.7

All Outpatients who developed fever were admitted and administered IV antibiotics. First line antibiotics included piperacillin-tazobactum or cefoperazone-sulbactum, along with amikacin. Second line empirical antibiotics were generally started if fever persisted for 48–72 hours and there was no clinical improvement. These consisted of a carbapenem or aztreonam. Vancomycin/teicoplanin were added at onset or later as per febrile neutropenia guidelines.7 Amphotericin was added if there was any suspicion of fungal infection based on clinical or X-ray findings, and empirically if fever and neutropenia persisted despite antibiotic therapy for more than five days.

After resolution of fever and if there was no obvious infection, IV antibiotics were changed to oral antibiotics that were continued for at least five days or till ANC recovered. Outpatients were discharged once oral antibiotics (amoxycillin-clauvulanic acid and levofloxacin) were initiated, while Inpatients remained in hospital till recovery of ANC.

The number of febrile neutropenic episodes, use of antibiotics, patterns of infection and mortality after completion of HiDAC were compared in the 2 groups. The SPSS statistical software (Chicago, IL, USA) was used for analysis.

Table 1.Characteristics of Outpatient and Inpatient neutropenic episodes.

Patients’ characteristics and outcome in the 2 groups is given in Table 1 and Figure 1. The relative risk of developing fever in Inpatients was 1.51 (CI 0.86–2.66, using the χ test). First line antibiotics cured fever in 88% (22/25) Outpatient febrile neutropenic episodes compared to 58% (14/24) Inpatients. Patients who remained hospitalized after chemotherapy had a higher incidence of culture positive fevers as compared to domiciliary neutropenic fever. There were 3 deaths in the study group (Figure 1).

In recent years, several reports have questioned the necessity of keeping patients in hospital after chemotherapy till full neutrophil count recovery.3–6 There is little published data of the safety of this approach in centers with limited resources.

Domiciliary management of low risk febrile neutropenia with oral antibiotics relies on an efficient health care infrastructure, where patients would be immediately admitted if their condition deteriorates. Lacking these facilities, we decided to admit all Outpatients who developed fever during neutropenia. Although current guidelines do not recommend antibiotic prophylaxis for neutropenia,7 new evidence shows benefit.8,9 In our limited resource setting prophylaxis with ciprofloxacin seemed advisable. Fluconazole is recommended for allogeneic bone marrow transplant patients with varying doses.10,11 We used fluconazole in 200 mg daily due to the low average weight of our patients.

Figure 1.Figure showing the outcome of neutropenic episodes.

In a recent editorial, Kern2 emphasized that improved antimicrobial prophylaxis (fluoroquinolones, aciclovir, fluconazole) together with much more effective supportive care algorithms on one hand, and professional risk assessment and appropriate infrastructure for follow-up on the other, are essential in discussing the successes of early discharge and outpatient management programs.

The one mortality in the Outpatient group was likely due to delayed presentation as the patient reported to hospital 24 hours after onset of fever and died within 18 hours of admission despite intensive resuscitation. It is, therefore, critical to ensure that patients understand the importance of following instructions to report to a medical center immediately on onset of fever or other symptoms suggestive of infection.

Of the 48-neutropenic episodes managed on a domiciliary basis, 23 (48%) did not require admission during their entire nadir. Of the other 25 neutropenic episodes that required re-admission, many hospital days were still saved by the early discharge policy.

The study shows that selected patients can be discharged and given domiciliary treatment safely, even with inadequate home care facilities. However, if the patient cannot reach the hospital immediately in cases of fever, early discharge should be avoided.