Occurrence of infections and its relationship to hemogram in patients with acute leukaemia.
Infection is still a major cause of death in patients with malignant blood diseases, especially in acute leukemia.,  About 80% of the patients with acute leukemia develop infections during the course of their disease and many of these patients have multiple episodes of infection. As the anticancer treatment is commenced; the host defences are altered and the patient readily becomes infected even with non bacterial opportunistic agents like fungi, viruses and protozoa.
Ninety patients having acute leukemia who were treated in the Tata Memorial Hospital and K. E. M. Hospital were included in the study. Their ages varied from five years and above. Samples of urine, blood and throat swabs were collected for microbiological examination at various intervals during the course of the disease. These samples were collected immediately after the diagnosis but before the onset of induction therapy, during leukopenia, in febrile conditions, after the complete course of induction therapy, after complete remission, during relapse and blastic crisis. In case of scanty growth and suspected contamination, urine cultures were repeated.
Throat swabs were collected from 15 normal individuals not taking any antibiotics, to study the normal flora.
The isolated organisms were identified by biochemical and morphological characteristics. The bacteria were identified according to the method described by Gradwohl and Baily and Scott. The fungi were identified according to the methods described by Beneke and Rogers. The in vitro antibiotic sensitivity of all the isolated organisms was done. Hemograms were done from all the blood samples.
One hundred and seventy three samples were collected from 90 patients with different types of acute leukemia, namely acute myeloid leukemia (49 patients), acute lymphoid leukemia (17 patients), acute erythro-leukemia (2 patients), acute monocytic leukemia (4 patients), acute monomyelocytic leukemia (9 patients), acute promyelocytic leukemia (1 patient), acute undifferentiated leukemia (4 patients) and chronic myeloid leukemia in blastic crisis (4 patients). The spectrum of bacterial and mycotic infections occurring in these patients was studied. It was observed that the occurrence of infections was influenced by the following major factors: (a) the amount of hemoglobin, (b) absolute white blood cell counts, (c) absolute neutrophil counts and (d) absolute blast cell counts in the peripheral blood.
An inverse relationship was seen between the amount of hemoglobin and susceptibility to infection in a given patient [Figure 1]. Thus, the incidence of infection was found to be maximum in patients with hemoglobin levels below 4.0 grams per cent whereas the incidence of infection was minimum in the group with hemoglobin levels above 10.1 grams per cent.
The incidence of bacterial infections was found to increase with white blood cell counts above or below the normal range. In the former condition, increase in cell counts was due to the presence of large numbers of immature cells. Fungal infections in the urinary tract and throat occurred mostly when white blood cell counts were below the normal range. No fungal infections were detected in the blood. The occurrence of infections was maximum at white blood cell counts of 1000/cmm and below [Figure 2].
[Figure 3] shows the relationship between bacterial and fungal infections and absolute neutrophil counts. The occurrence of infection was comparatively high at absolute neutrophil counts below 1000/cmm. A good number of infections also occurred at higher absolute neutrophil counts.
The occurrence of infections increased with an increase in absolute blast cell counts in the peripheral blood. [Figure 4].
[Table 1] shows the incidence of infection with different types of organisms. E. coli was the most common organism isolated in the urinary tract. In blood, Staphylococci and Pseudomonas were commonly encountered. Klebsiella and Candida were commonly isolated from the throat.
No Candida was isolated from any of the throat swabs collected from normal individuals, not taking any antibiotics. This clearly indicated that the bacterial flora had an inhibitory action on Candida. Pseudomonas was the most important infectious agent, because although the percentage of occurrence of Pseudomonas infection was as low as 11%, the percentage mortality with this organism was as high as 87%.
The in vitro antibiotic sensitivity of various organisms causing infections is shown in [Table 2]. Most of the bacteria were found to be sensitive to gentamicin (58.91%) and polymyxin B (64.81%), while all the Caadida (66 samples-100%) were found to be sensitive to both amphotericin-B and nystatin.
The high incidence of morbidity and mortality in patients with acute leukemia is mainly due to secondary bacterial and mycotic infections. This can be attributed to several factors that impair host defense mechanisms in these patients. Firstly, the number of mature granulocytes which play a role in host defense mechanism is very low. Secondly, the induction therapy given to destroy the leukemic cells results in further reduction of the mature granulocytes.
The results in the present study show a preponderance of infections in patients with low hemoglobin levels. Hemoglobin level generally reflects the health of the patient and is inversely proportional to the extent of the disease. The high incidence of infections observed in patients with either leucopenia or high white blood cell counts suggest that the increased susceptibility was due to the reduced number of mature cells in the circulation.
There is a correlation between the degree of neutropenia and susceptibility to severe infection. Changes in neutrophil counts during the period of infection influence the outcome of treatment. The death rate in the group of patients having an absolute neutrophil count of less than 100 cells per cmm was comparatively much higher than in patients with an absolute neutrophil count of more than 1000 cells per cmm. These results correlate well with those of Chang et al who showed that the death rate was high at neutrophil counts of less than 100/cmm and it was comparatively lower at terminal neutrophil counts above 1000/cmm. A good number of infections also occurred at higher absolute neutrophil counts. This may be attributed to the dysfunction of the mature neutrophils. The high incidence of infection at high blast cell counts probably suggests that the blast cells are unable to act like the normal neutrophils. Both the disease and its treatment may make patients vulnerable to severe infections. They also depress the ability of the host to evoke an acute inflammatory reaction.
Many workers have observed that lack of symptoms and signs of infection occurs most frequently in neutropenic patients.,  Also most of the aminoglycoside antibiotics, like gentamicin, become ineffective at low circulating granulocyte levels. By contrast, they have an increased frequency of dissemination of infections as evidenced by the sharply increased incidence of bacteremia. Therefore, there should be an immediate administration of broad spectrum antibiotics when there is a clinical picture of infection. Empirical antibiotic therapy should include a combination of antibiotics active against both Gram positive and Gram negative organisms, particularly against Pseudomonas isolates which are commonly encountered during neutropenia. It had previously been demonstrated that delay of therapy in neutropenic patients who develop Gram negative bacterial infections is associated with a high mortality rate.
We are thankful to Dr. C. K. Deshpande, Dean, G.S. Medical College and K.E.M. Hospital for allowing us to publish the work done in this hospital. We are also thankful to Dr. D. J. Jussawalla, Director, Tata Memorial Hospital, for allowing us to use the material from their patients for this work.