Epidemiology and Clinical Characteristics of Mucormycosis in Patients with Leukemia; A 21-year Experience from Southern Iran

Abstract

Objectives:

To determine the epidemiological aspect of mucormycosis, the nature of malignancies complicated by mucormycosis, the initial site of involvement and the subsequent outcome.

Methods:

This was a cross-sectional study which was performed by reviewing the medical records of 95 patients with leukemia complicated with biopsy-proven mucormycosis admitted to the educational hospitals affiliated with Shiraz University of Medical Sciences over a 21-year period. We recorded demographic information including age and sex and disease characteristics such as type of leukemia, site of involvement, paraclinical findings at the time of admission and the outcome of the patients. The incidence of mucormycosis in leukemia was determined by identifying the number of leukemia patients diagnosed within the last 17 years.

Results:

The male to female ratio was 2.39:1 in of 95 patients studied. The overall incidence rate of mucormycosis was 4.27 per 100 leukemic patients in last 17 years which showed a decreasing trend from 2001 to 2011. The most frequent type of leukemia was acute myelogenous leukemia (AML) which was found in 58 patients (61.5%). The most common site of initial tumor involvement was sinonasal (90.16%). The mortality rate was about 54%, compared to the mortality rate of about 43.24% in patients with best prognosis of AML.

Conclusion:

The incidence of mucormycosis in leukemia showed a decreasing trend in our country and its recent incidence is comparable to that of other regions. The best preventive method against this lethal infection is to modify and control the environment which reduces the risk of exposure to air-born fungal spores.

Introduction

Zygomycosis  is   an   invasive  fungal   infection which is primarily caused by fungi belonging to the Mucoraceaefamily. As in other  invasive fungal diseases, impairment of the immune  system is the most important predisposing factor. Patients at risk usually have compromised defence mechanisms, mainly caused by diabetes mellitus,  hematological  diseases like leukemia or lymphoma, i.v. drug abuse or cytotoxic therapy.

Although there are reports of zygomycosis in otherwise healthy patients [1]. Mucormycosis is a rare invasive fungal infection seen most often in patients with  haematological  malignancies, particularly  in the neutropenic phase [2]. After germination of the spores, the resulting hyphae proliferate and  as the Mucorales have a particular predilection for blood vessels, angio-invasive disease  is common, often leading to thrombosis and infarction of surrounding tissue. The type, extent and severity of the disease depend on the host defences and different risk factors [3]. In leukemia patients, zygomycosis usually leads to a delay or interruption of anti-leukemic therapy, thus  increasing mortality  not only through the fungal infection itself, but also from the inability to adequately treat the hematologic disease [1].

The diagnosis of zygomycosis is rarely suspected and antemortem diagnosis is made in only 23-50% of cases [4]. Early medical and surgical treatment could prevent further dissemination but some patients may be cured by surgical excision and amphotericin [5]. The association of this infection with haematological malignancies has not been adequately addressed in the  literature. The aim  of this retrospective study was to analyze the experience of our center about mucormycosis  in  haematologic  malignancies. We conducted a 21-year retrospective review of all cases that  hospitalized in 3 large Shiraz university hospitals as a referral center in southern  Iran. The results were used to assess trends in the epidemiology of this infection, study the nature of malignancies complicated by  mucormycosis, the initial site  of involvement, the subsequent outcome of the victims and analyze their laboratory findings at the time of admission.

Materials and Methods

Study population

 This  was  a   retrospective  cross-sectional  study being performed  in Nemazee, Shahid Faghihi and Khalili hospitals  affiliated  with  Shiraz  University hospitals, principal referral centers in southern Iran, over a period of 21 years from April 1990 through March 2011. We included all the patients who were pathologically diagnosed to have mucormycosis and leukemia who were admitted to our centers during the study period. We excluded those with underlying diseases other than  leukemia including diabetes mellitus (DM). Of 162 mucormycosis cases 95 (58.6%) had leukemia as their underlying disease. The number  of patients with leukemiain southern  Iran was determined by the health centers during the past 17-year period from1995 to 2011. The study protocol was approved by Shiraz University of Medical Sciences institutional review board (IRB) and medical research ethics committee. As this was a retrospective analysis of medical charts, no informed written consents were obtained from the patients.

Study protocol

Evaluation  at  presentation   included   a  detailed history, and  otorhinolaryngologic, ophthalmic  and neurologic examinations to assess the extent of the disease. Laboratory  investigations  included  white blood cell count (WBC), hemoglobinlevel, blood urea nitrogen (BUN), blood sugar, serum bicarbonate and arterial PH. Diagnosis was made on histopathological examination and KOH preparation of biopsy specimens obtained  from  the  nasal cavity and/or paranasal  sinuses  and  the  palate. An  orbital  fine needle aspiration cytological (FNAC) study was done in some patients  when histopathological diagnosis from other sites was equivocal. Computerized tomographic  scans of the paranasal sinuses, orbits and brain were obtained to assess the extent of disease. All these information were obtained from the medical charts by reviewing and entering into a standard data gathering form.

The  treatment  strategies were similar  in  all the patients. All the patients were treated with systemic Amphotericin B as soon as the diagnosis of mucor was established, along with treatment to stabilize the underlying metabolic derangement. After a test dose of 1 mg of amphotericin B in 100 ml of normal saline, 0.7 mg/kg/day of amphotericin  B was given over 6 hours.  All the  patients  underwent  local resection of the lesions until a safe margin was reached. The outcome of treatment group was evaluated in terms of treatment success and treatment failure. Treatment success was defined as a disease-free, stable patient with controlled  metabolic status. Treatment  failure was defined as progression of disease to a more advanced stage, worsening general condition or mortality due to the infection. We also recorded epidemiological characteristics, the  site  of  infection  and  type  of leukemia and its duration and characteristics.

Statistical analysis

 The statistical package for social science, SPSS for Windows, Version 15.0 (SPSS Inc., Chicago, IL, USA) was used for data analysis. Data are reported as mean ± SD and proportions as appropriate. A p-value less than 0.05 were considered statistically significant.

Results

A review of hospital records identified 162 cases of mucormycosis admitted to Shiraz University hospitals, during the last 21 years from April 1990 through  March 2011, of which 95 (58.6%) patients had leukemias their underlying disease of which 67 (70.5%) were males and 28(29.5%) were females(Male to  Female ratio: 2.39:1). The age of patients  with leukemia and  mucormycosis ranged from  6 to  80 years with mean of 30.08±16.25 years. Figure 1 shows the number of mucormycotic leukemia patients admitted tour departments during the study period.

Fig. 1.

The number of leukemia patients with mucormycosisadmitted toShirazUniversity hospitals from 1990 to 2011

The  number  of  patients  increased significantly in years 1999 and 2000. The annual incidence rate of mucormycosis in leukemia patients during the last 17 years was found to  be  significantly higher  (p<0.001)  in  1999 and 2000 compared to other years (Figure 2). An overall incidence rate  of mucormycosis was 4.27 per  100 leukemia patients  in  a  17-year period  from  1996 to  2011. Distribution  of  mucormycotic leukemia patients according to the month  of onset is shown in Figure 3, without noticeable seasonal pattern (p=0.125).The most frequent type of leukemia was acute myelogenous leukemia (AML) that was found in 58 (61.5%) patients followed by acute lymphogenous leukemia (ALL) in 25 (26.3%), chronic myelogenous leukemia (CML) in 6 (6.3%), chronic lymphogenous leukemia (CLL) in 2 (2.1%), biphenotypic leukemia in 1 (1.06%), hairy cell leukemia in 1 (1.06%) and unknown types of leukemia in 2 (2.1%) patients.

Fig. 2.

Annual incidence rates of mucormycosis in leukemic patients in southern Iranfrom1996 to 2011.

Fig. 3.

The number of mucormycoticleukemic patients admitted to Shiraz University hospitals based on the month of onset from1990 to 2011

The initial site of involvement in 83 (88.6%) patients was sinus, nose or both, and in 2 (2.1%) case shard palate was involved. The tonsils, lung, vagina, and arm skin were affected in1 (1.06%) patient each. The laboratory findings of the patients are summarized in Table 1. We found that 50 (52.6%) patients had WBC count lower than 3.5×103/μlit while in 30 (31.57%) patients WBC count was more  than 10×103/μlit. Hemoglobin level in 20 (21.3%) cases was lower than 7gr/dlit, in 60 (65.58%) was between 7 and 10 gr/ dlit and in 15 (14.73%) higher than 10gr/dlit. Serum pH in 28 (29.47%) cases was higher than 7.45 while in 3 (3.15%) patients it was lower than 7.35. Serum bicarbonate was higher than 24mg/dlit in 9 (9.47%) and lower than15 mg/dlit. in 6 (6.31%) patients.

Table 1.

Laboratory findings of leukemiamucormycotic patients in southern Iranfrom1990 to2011.

Laboratory test (measure unit)	Min	Max	Mean
White blood cell count (103/μlit)	0.1	228.3	19.26
Blood urea nitrogen (mg/dlit)	5	101	19.18
Hemoglobin level (gr/dlit)	4	13	8.05
Fasting blood sugar (mg/dlit)	55	433	140.77
Serum PH	7.22	7.57	7.43
Serum bicarbonate (meq/lit)	6.5	31.6	21.37

A total number of 54(56.8%) patients died of which 43% had AML with mucormycosis. About 75% had ALL with mucormycosis and  75% had  CML with mucormycosis. Other fatalities included one patient with  CLL and  one  case  with  unknown  type  of leukemia.

Our analysis showed that the patients’ age, sex and laboratory findings had no effects on the prognosis of these patients. All patients received medical treatment with amphotericin B early after suspicious diagnosis with  mucormycosis,  while 87 (91.5%)  them  also underwent surgical intervention. The time lag between onsets of symptoms  referable to  zygomycosis and starting amphotericin B administration ranged from 3 to 85 days with a median of 9 days. The antifungal dose ranged from 0 to 3300 mg with median 1000 mg and duration of treatment being 35 days which ranged from1 to 92 days. The median time lag between onset of symptoms and surgical management was 12 days with a range varying from 7 to 45 days.

Discussion

Invasive fungal  infections  caused  by  yeasts and moulds   are   frequent   problems   in   hematologic patients [1]. Zygomycosis is the third most common invasive fungal infection  next  to  candidias  is and aspergillosis that accounts for 8.3-13% of all fungal infections  discovered at  post  mortem  in  patients with hematologic disorders [4]. This is evidenced by an increase in zygomycosis from 8 to 20 per 100,000 admissions between 1989 to1993 and  1994to1998. However in hematopoietic stem cell transplantation patients  the  rate  was 10/4020 (0.25%)  [6].  Three independent publications  reported  rates  of  8.9%, 3.2% and 4.3% in HSCT patients, all attributed the sharp rise in voriconazoleuse in late 2002 [7-9]. In our  study, a significant increase was found  in the number of patients with haematologic malignancies superimposed by mucormycosis in 1999 and 2000. All our patients were diagnosed before death, and overall the incidence rate of mucormycosis in leukemic patients was 4.27 per 100.Considering the patients with undiagnosed mucormycosis before death, the virtual incidence rate of mucormycosis in leukemic patients in our country might be higher than other regions. But the annual incidence rate of mucormycosis was not the same. The highest incidence rate was seen in 1999 followed by a decreasing trend    in ensuing decade. Following increase in the incidence of mucormycosis in 1999, attempts were made to control and modify the conditions in haematological wards to reduce the risk of exposure to air-born  spores. These included excluding flowers and live plants from these wards, continuous  isolation  of  such  neutropenic  patients who are at higher risk for these fungal infections, wearing masks and  monitoring  of air quality and condition. The incidence of mucormycosis was then significantly  decreased  by  taking  such  measures and conducting appropriate interventions so that it became consistent with that of developed countries in subsequent decade.

The association of type of leukemia and mucormycosis has not been adequately addressed in the literature but in our study the most common neoplasm complicated by this fungal infection was AML followed by ALL.

Various studies yielded different results about the influence of age and sex on mucormycosis. Bitaret et al. believed that strict comparison of age distributions was not  possible in  different  studies  [10].  But  it seems that sex is an important  predisposing factor for mucormycosis. In a recent study conducted  by Jagarlamundi et al. it was shown that all patients with acute leukemia and mucormycosis the male sex was dominant  (M:F=2.5:1) where they age of patients ranged from 6 to 66 ( mean age 36 years) [11]. In our study, themselves were also dominant  (M:F=2.38:1) and the patients aged from 6 to 80 ( mean age 30.08 years). Considering that diabetic female patients were older  (mean  age 61 years) and  acquire  infections more often than males [12]. It seems that gender and age distributions  confer most important  impact on underlying diseases in patients with mucormycosis where  diabetics  are older  females  and  leukemia involve younger male patients. Rhinocerebralzygomycosis   is   the   most   common presentation and almost always associated with hyper glycemia and metabolic acidosis [13]. It is reported to  be  the most  frequent  form  of  mucormycosis in  several studies  [14-16]. Patients  with leukemia have traditionally been considered to represent the majority  of  cases with  pulmonary  mucormycosis [16] and in some studies pulmonary mucormycosis is the most prevalent form in patients with leukemia [2]. A retrospective study by Pagano et al. identified 59 patients with haematological malignancy and mucormycosis. The most common presentation was pulmonary mucor, which accounted for 64% of cases of which  24%, involved orbital  and  maxillofacial areas [2]. The majority of our cases presented with rhinocerebral  mucormycosis  and  only one  patient had pleural involvement.

The main  cause of this infection is probably the prolonged   and   profound  neutropenia   secondary to the myeloablative treatments  for the underlying hematologic malignancy [2]. The importance of neutropenia  as a contributing  factor to infection in patients with hematologic malignancies has already been demonstrated [2].Consistent with other studies we found that majority of our cases had neutropenia or leukocytosis which supports that neutrophil dysfunction is the major risk factor for developing fungal infections in haematologic malignancies. Several researches show a preponderance of infections in patients with low hemoglobin levels and anemia such as thalassemia or aplastic anemia as predisposing factors for mucormycosis [17-19] which is inversely proportional  to the extent of the disease. Nearly all of our patients were anemic at the time of acquiring infection.

The widespread use of steroids had also resulted in  an  increased incidence  of zygomycosis in  this population of patients who had chronically or acutely been treated  with this class of drugs [19-21]. The mechanism by which the corticosteroids enhance susceptibility to  zygomycosis is probably two fold. First, steroids suppress the normal inflammatory cell response and second they may induce a diabetic state [19-21].

Invasive zygomycosis is  simply  dismal  news  for patients as well as treating physicians. Indeed, despite recent  medical advances, this  aggressive fungal infection still  carries  poor  prognosis.  Since deep mucormycosis  encompasses  many syndromes, the mortality rate varies greatly from 33.3% in a Korean study [22] to a worse rate of 63% in an Italian report [2] to a staggering 96% rate in regard to disseminated form [13]. This extreme variation in mucormycosis mortality  rate  can be  explained by many  factors, including early diagnosis, site of the infection, patient's immune status, correction of other co- morbid factors, and among others the type of therapy instituted. In our study all patients were treated with amphotericin   B  and deoxycholate.  Lipid  soluble form of amphotericin B was not used because of its higher cost and lack of availability. Only 8 patients did not undergo surgical treatment and patients with more surgical interventions had better outcome. The overall 56.8% mortality  rate  was reported  in  our study which was consistent with other reports [12-14]. Mucormycosis occurred  in all of our  patients after receiving chemotherapy that had corticosteroids in their chemotherapy protocol. It seems that about 44% of our patients were in diabetic state without any history of diabetes before developing leukemia. Because mucormycosis is a highly lethal infection with mortality rates of   50% or more and despite aggressive management  even  in  the  best medical centers, the prevention and lowering the incidence of this infection is a priority for immune-compromised patients such as those with malignancies (especially hematologic), diabetic ketoacidosis and AIDS, who seem to be at higher risk for this infection. Considering that the most common method of transmission of this infection is via inhalation of spores that are mainly present in the soil, air, atmosphere of hospital wards and decaying materials, taking appropriate measures to control and  modify the environment that reduce the risk of exposure to fungal spores seems to be the best preventive approach for this lethal infection.

Conflict of Interest: None declared.