Infectious complications during therapy of acute leukaemia in Saudi Arabia

Journal

of Hospital

Infection

Infectious

(1989)

14, 209-215

complications leukaemia

N. S. El Saghir,

during therapy in Saudi Arabia

of acute

A. D. Hall*, C. R. Santhosh-Kumar R. C. Zeitany*

and

Departments of Medicine and Pharmacy, * King Khalid University Hospital, King Saud University College of Medicine, Riyadh 11472, Saudi Arabia Accepted

for publication

2 February

1989

Summary: In 40 febrile neutropenic episodes during the induction and consolidation chemotherapy of acute leukaemia in Riyadh, 51% of organisms causing septicaemia were gram-negative, 26% gram-positive, 8% anaerobes and 15% fungi. In 21 (52%) febrile episodes there were pulmonary infiltrates; of the 12 where aetiology was known, six were due to fungi. Pulmonary infiltrates progressed to adult respiratory distress syndrome and death in nine instances. There was no significant occurrence of parasitic and tropical infections. The results show that the pattern of infections, during therapy of acute leukaemia in developing countries, may have important differences when compared with western centres. Empiric amphotericin B may need to be introduced at an earlier stage in patients with persistent fever or progressive pulmonary infiltrates.

Keywords:

Acute

leukaemia;

septicaemia;

pneumonia;

fungal

infection.

Introciuction Infection is the commonest and most serious complication in acute leukaemia during periods of neutropenia (Bodey, Bolivar & Fainstein 1982; Pizza et al., 1984). Th e risk of serious infection increases markedly as absolute neutrophil counts fall below 1000 mmp3 and serious complications may arise if infection is not treated immediately (Bodey et al., 1966a). Empiric treatment of febrile episodes is an essential part of the management of such patients (Young, 1986). In more than 50% of instances microbiological evidence of infection is not forthcoming (Dunbile, 1988). Superinfections with fungi and other opportunistic pathogens are mainly seen in prolonged neutropenia, or during prolonged use of broad spectrum antibiotics, (Mirsky & Cuttner, 1972; Pizza et al., 1982) or repeated courses of chemotherapy (Robertson & Larson, 1988). Aggressive chemotherapy for acute leukaemia is currently given in many hospital centres in third world countries. Infectious complications in these Correspondence Michigan 48162, O195-6701/89/070209+06

to: Dr USA.

Nagi

S. El

Saghir,

Fordson

$03.00/O

aWedical

Clinic

5641,

Schaefer

Q 1989 The Hospital

209

St.,

Dearborn,

Infection

Soaety

210

N. S. El Saghir

et al.

countries may vary from those seen in the western world, because of different environments and pathogens, or susceptibility to infection. We report the infections seen during treatment of acute leukaemia at the King Khalid University Hospital (KKUH) in Riyadh, Saudi Arabia. Our hospital opened in 198 1 and has had a Haematology-Oncology service since 1984.

Patients

and methods Thirty adult patients with acute leukaemia were treated at KKUH in Riyadh between September 1984 and July 1987. Twenty two patients had acute non-lymphocytic leukaemia (ANLL) and eight patients acute lymphoblastic leukaemia (ALL). We have described our patient characteristics elsewhere, (El Saghir et al., 1987a; El Saghir et al., 1987b). Of the 27 available patients, ages ranged between 14 and 74 years with a median of 37 years. There were 19 males and eight females. Eighteen patients were Saudi arabs, two Filipinos, two Egyptians, and one each from Eritrea, Sudan, Pakistan, Sri-Lanka and India. Patients were treated with chemotherapy regimens consisting of daunorubicin, cytosine arabinoside and 6 thio-guanine for ANLL, or duanorubicin, cytosine arabinoside, cyclophosphamide, methotrexate vincristine, L-asparaginase, and intrathecal methotrexate for ALL. All patients had cultures taken from blood, urine, stool, and swabs from anterior nares, throat, ears, axillae and perianal areas on admission and weekly thereafter, and during episodes of fever. At least two sets of blood cultures were taken before institution of empiric therapy. All neutropenic patients (neutrophil count < 1000 mme3) were placed in protective isolation, with reverse barrier nursing and strict hand-washing techniques. Mouth washes with nystatin and 1.5% phenol solution (‘Chloraseptic’) were given to all patients throughout their hospital stay. The 17 patients who had Hickman catheters inserted received prophylactic intravenous vancomycin 500 mg six hourly, beginning one hour prior to catheter insertion and continuing for three days. Fever was defined as a temperature of > 38.5”C or two readings of > 38°C within a 24-h period, not obviously related to the use of blood products or amphotericin B. A combination of piperacillin and gentamicin in the first part of the study (19 patients), or ceftazidime alone (De Pauw et al., 1983, Pizza et al., 1986) in the later part of the study, (8 patients) were used as initial empiric therapy.-Three patients could not be evaluated since they left B was added if hospital against medical advice. Empiric amphotericin unexplained fever persisted for more than seven days during the initial part of the study, or earlier during the latter part, or if fungal infection was afterwards suspected. Aminoglycoside dosage was calculated according to Sawchuk et al. (1977). Antibiotic therapy was adjusted in the light of culture results or clinical course.

Infections

in acute

During treatment, data was collected database and spreadsheet program.

211

leukaemia

and recorded

into a computer

Results

A total of 27 patients were available for analysis. They had 42 admissions during their induction and consolidation chemotherapy, 40 of which were for neutropenia with fever. Neutropenia was present on admission in nine patients and was seen during therapy in all patients. Average duration of neutropenia was 23 days. (Range 2 to 52 days.) Septicaemias Septicaemia was documented by positive blood cultures in 31 of the 40 febrile episodes (78%). Two organisms were found in four episodes and three in two; giving a total of 39 isolates. Twenty isolates were gram-negative aerobic bacteria (51%), ten were gram-positive aerobic bacteria (26%), three were anaerobic bacteria (8%) and six were fungi (15%). Three of the fungaemias occurred with documented septicaemias. Septicaemias and frequencies of isolated organisms are detailed in Table I. Pulmonary infiltrates Pulmonary infiltrates occurred in 21 febrile episodes (53%). The aetiology was identified in 12; in five the same organism grew from blood and sputum

Table

I. Causative

organisms

of 31 episodes

of septicaemia No.

Organisms

(%)

bacteria

20 (51%) 8 5 4 2 1

Aerobic Gram-positive bacteria Staphylococcus epidermidis Staphylococcus aureus Streptococcus spp.

10 (26%) 6 2 2

Aerobic Gram-negative Escherichia coli Klebsiella spp. Pseudomonas spp. Enterobacter spp. Salmonella spp.

Anaerobes Bacteroides fragilis Propionobactevium

spp.

Fungi Candida

albicans

Number of febrile episodes: 40 Tom1 number of isolates: 39

3 (8%) 1 2 6 (15%) 6

212

N. S. El Saghir

et al.

simultaneously (gram-negative bacteria in two, gram-positive bacteria in one and fungi in two). In three the sputum grew Mycobacterium tuberculosis in the presence of chest radiographs suggestive of pulmonary tuberculosis, and in two patients fungi (Aspergillus terreus and Trichosporon beigelii) were cultured from sputum, from patients with typical chest x-ray features of fungal infection, i.e. fungus balls or cavitation (El Saghir et al., in press; Santhosh-Kumar et al., 1989). In two episodes lung biopsy yielded Catidida albicans. The aetiology of the pulmonary infiltrates was not clear in the remaining nine instances. Pneumonia progressed to adult respiratory distress syndrome (ARDS) and death in nine patients; two had simultaneous candidaemia, three had gram-negative septicaemia, and two had C. albicans on lung biopsies. Urinary tract infections There were ten urinary tract infections; six were due to gram-negative organisms and four due to fungi, in two of whom there was also candidaemia. Other infections Only one parasitic infestation (Strongyloides stercoralis) was seen. There was one instance of documented scrotal abscess (Staphylococcus aweus), one perianal abscess (Enterobacter cloacae), one Bartholin gland abscess (E coli), one episode of oesophageal candidiasis and two episodes of herpes simplex infection. Discussion

Our results show several differences when compared with western experience in the patterns of infection seen during neutropenia following chemotherapy of acute leukaemia. We had a higher rate of septicaemia per febrile episode (77%), and significant numbers of infections with multiple organisms and fungi. A comparison of the percentages of blood culture isolates in our neutropenic acute leukaemia patients, with those reported from the EORTC trial III (Klastersky et al., 1986) in neutropenic cancer patients, shows an equivalent percentage of gram-negative organisms, slightly lower percentage of gram-positive organisms, and a significantly greater number of fungal organisms (PC O-05) and anaerobes (PC 0.01) (Table II). Many patients with.pneumonia progressed to ARDS and died despite empiric treatment with broad spectrum antibiotics and amphotericin B. These complications had adversely affected the remission rates in our patients (El Saghir et al., 1987a; b) despite the use of standard chemotherapeutic regimens. Whether our patient population has a lower tolerance of aggressive chemotherapy or a different pattern of colonisation and spread of microorganisms is not known. However, it is our impression,

Infections Table

II.

Comparison patients

of blood

and

culture the EORTC

in acute

leukaemia

213

isolates between this series of neutropenic trial III of neutropenic cancer patients.

KKUH-Riyadh 1984-1987

EORTC

Trial 1982

III

leukaemic

Statistical* Significance

NO.

(%)

NO.

(%I

Gram-negative organisms

20

(51)

57

(57)

NS

Gram-positive organisms

10

(26)

57

(37)

NS

3

( 8)

0

( 0)

P
-

( 5)

PCO.05

Anaerobes Fungi Total *Chi-Square

6 39

(15) 9;

analysis

and that of other oncologists in this area, that patients in this part of the world do not tolerate aggressive chemotherapy as well as patients in the West. All our patients were either Saudi arabs, or migrants from other third world countries. It is possible that the true incidence of fungal infections in our patients is even higher than found here, because fungaemia may be documented in only 25 to 30% of disseminated deep tissue fungal infections (Bodey, 19666). Fungal infections occur in neutropenic patients but are normally considered to be rarely rapidly lethal, as compared with gram-negative infections. Empiric coverage for these organisms is usually recommended only in cases of persistent fever, or on clinical suspicion after a period of five to seven days, or in the presence of rapid clinical deterioration (Pizza et al., 1982; Young, 1986; Robertson & Larson, 1988). By contrast, our results show that fungal infections may be rapidly fatal in this group of patients. The progression of pneumonias from minimal basal infiltrates into diffuse bilateral pneumonia and ARDS, together with a significant number of fungal isolates, has led us to a policy of instituting empiric amphotericin B as soon as patients develop minimal basal infiltrates and show deterioration of arterial blood gases. On the other hand, as in western centres (Bodey et al. 1982), our data show that gram-negative organisms are the commonest and most serious complications that occur in these patients and require immediate empiric treatment at the onset of fever. It is of interest to note that parasitic infections and other endemic or epidemic tropical diseases in Saudi Arabia, such as brucellosis, malaria, or salmonellosis caused no significant morbidity nor mortality in our severely compromised patients. We conclude that for empiric antibiotic therapy during febrile

214

N. S. El Saghir

et al.

neutropenia in acute leukaemia in Third World countries, amphotericin B may need to be added empirically after no longer than 72 h of persistent fever while receiving broad spectrum antibiotics, or in patients showing deterioration of arterial blood gases or progression of lung infiltrates. Further reports from Third World centres are required to confirm our results, and to improve treatment outcome. The authors wish to thank Prof. Robert E. Mass, Dr. John McIver, Dr. Steven Wright, Dr. D. S. Ajarim, Dr. K. E. Higgy, Dr. K. Al Khairy, Dr. S. A. Al Mashhadani, Dr. Essam Fawzy and Dr. A. K. Al Momen for their valuable help in the management of the patients. They also thank Dr. A. M. Kambal and Dr. N. Sulaiman from the Microbiology Unit for their excellent support.

References Bodey,

G. P., Buckley, M., Sathe, Y. S. & Freireich, E. J. (1966~). Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Annals of Internal Medicine 64, 328-340. Bodey, G. P. (19666). Fungal infections complicating acute leukemia. Journal of Chronic Diseases 19, 667-687. Bodey, G. P., Bolivar, R. & Fainstein, V. (1982). Infectious complications in leukemic patients. Seminars in Hematology 19, 193-226. De Pauw, B. E., Kauw, F., Muytjens, H., Williams, K. J. & Bothof, T. (1983). Randomized study of ceftazidime versus gentamicin plus cefotaxime for infections in severe granulocytopenic patients. Journal of Antimicrobial Chemotherapy 12 (Suppl. A), 93-99. Dunbile, M. J. (1988). Stopping antibiotic therapy in neutropenic patients. Annuls of Internal Medicine 108, 289-292. El Saghir, N. S., Mass, R. E., Hall, A. D., Ajarim, D. S. & Higgy, K. S. (1987~). ANLL in Saudi Arabia: Increased frequency of M4 and MS types and high rate of serious and fungal infections during chemotherapy. Proceedings of the 29th Annual Meeting of the American Society of Hematology 1987. Blood 70 (Suppl. l), 760 (Abstract). El Saghir, N. S., Mass, R. E., Hall, A. D. & Fawzy, E. (1987b). Hoelzer’s regime for the treatment of acute lymphoblastic leukemia (ALL) in Saudi Arabia. Proceedings of the 29th Annual Meeting of the American Society of Hematology 1987. Blood 70 (Suppl. l), 760 (Abstract). El Saghir, N. S., Al-Hedaithy S. S. A., Hawass, N., Fawzy, E. & Hall, A. D. Pulmonary aspergillosis due to Aspergillus terreus var. africanus in acute leukemia. Annals of Saudi Medicine (in press) Klastersky, J., Glauser, M. P., SchimpE, S. C., Zinner, S. H. & Gaya, H. (1986). Prospective randomized comparison of three antibiotic regimens for empirical therapy of suspected bacteremic infection in febrile granulocytopenic patients. Antimicrobial Agents and Chemotherapy 29, 263-270. Mirsky, H. S. & Cuttner, J.. (1972). Fungal infections in acute leukemia. Cancer 30,348-352. Pizzo, P., Robichaud, K., Gill, F. & Witebsky, F. (1982). Empiric antibiotic and antifungal therapy for cancer patients with prolonged fever and granulocytopenia. American Journal of Medicine 72, 101-l 11. Pizza, R. A., Commors, J., Cotton, D., Gress, J., Hathorn, J., Hiemenz, J., Longo, D., Marshall, D. & Robichaud, K. J. (1984). Approaching the controversies in antibacterial management of cancer patients. American Journal of Medicine 76, 436-449. Pizza, P. A., Hathorn, J. W., Hiemenz, J., Browne, M., Commers, J., Cotton, D., Gress, J., Longo, D., Marshall, D., McKnight, J., Rubin, M., Skeleton, J., Thaller, M. &Wesley, R. (1986). A randomized trial comparing ceftazidime alone with combination antibiotic

Infections

in

acute

leukaemia

215

therapy in cancer patients with fever and neutropenia. New EnglandJournal of Medicine 315, 552-528. Robertson, M. J. & Larson, R. A. (1988). R ecurrent fungal pneumonias in patients with acute nonlymphocytic leukemia undergoing multiple courses of intensive chemotherapy. American Journal of Medicine 84, 233-239. Santhosh-Kumar, C. R., Al-Hedaithy, S. S. A., El Saghir, N. S. & Ajarim, D. S. (1989). Cavitating pneumonia due to Trichosporon beigelli. Journal of Infection 19, 65-68. Sawchuk, R. J., Zaske, D. E., Cipolle, R. J., Wargin, W. A. & Strate, R. G. (1977). Kinetic model for gentamicin dosing with the use of individual patient parameters. Clinical Pharmacology and Therapeutics 21, 362-369. Young, L. S. (1986). Empirical antimicrobial therapy in the neutropenic host. New England Journal of Medicine 315, 580-581.