Hepatic Involvement in Pediatric Patientswith Paracoccidioidomycosis: A Clinicaland Laboratory Study

Giselle de Melo Braga • Gabriel Hessel •

Ricardo Mendes Pereira

Received: 28 August 2012 / Accepted: 22 July 2013 / Published online: 6 August 2013

� Springer Science+Business Media Dordrecht 2013

Abstract The liver is one of the organs most affected

by paracoccidioidomycosis, a systemic mycosis ende-

mic in some Latin American countries. The majority of

articles focused on adult populations and failed to

describe any detailed experience of liver abnormalities

in pediatric patients. Therefore, the aim of this study

was to describe the frequency and characteristics of

liver involvement in children with paracoccidioido-

mycosis. This study comprised 102 patients less than

16 years of age (median 104.3 months) diagnosed with

paracoccidioidomycosis from 1980 to 2010. Diagnosis

was established by the identification of fungus. Forty-

one patients had liver involvement. The main clinical

features were generalized lymph node enlargement

(39/41), weight loss (34/41) and fever 32/41). Approx-

imately, one-third of the patients had jaundice. Patients

with hepatic involvement were younger. A predomi-

nant elevation of canalicular enzymes occurred. There

was a statistically significant difference in albumin

(p \ 0.001) and hemoglobin (p = 0.002) values

between patients with and without liver involvement,

and the lowest values were found in the former group.

Cutoff levels of albumin (\3.05 g/dL) and hemoglobin

(\9.2 g/dL) can be used to infer hepatic involvement.

Hypoalbuminemia (median 2.4 g/dl) is more severe in

patients with hepatic involvement and may indicate a

worse liver function or complication of the disease

(intestinal lymphangiectasia). Deaths (6) occurred

only among patients with liver involvement. Particular

clinical and laboratory characteristics are present in

pediatric patients with hepatic involvement. Younger

patients and those with severe hypoalbuminemia are

more likely to present liver involvement by Paracoc-

cidioides brasiliensis.

Keywords Paracoccidioidomycosis �Paracoccidioides brasiliensis � Liver � Pediatrics �Children � Serum albumin

Introduction

Paracoccidioidomycosis (PCM) is the eighth leading

cause of death from chronic infectious and parasitic

diseases with the highest mortality rate among

systemic mycoses in Brazil [1]. It is a systemic

mycosis caused by a thermally dimorphic fungus [2]

that occurs only in Latin America. Brazil has the

highest incidence of the disease [3, 4]. The etiological

agent of PCM, Paracoccidioides brasiliensis

(P. brasiliensis), is a cryptic species complex. It is

composed of at least four different phylogenetic,

potentially human pathogenic lineages (S1, PS2 and

G. de Melo Braga (&)

Center for Investigation in Pediatrics, UNICAMP School

of Medicine, Campinas, Sao Paulo, Brazil

e-mail: gimed38@gmail.com

G. Hessel � R. M. Pereira

Department of Pediatrics, UNICAMP School

of Medicine, Sao Paulo, Brazil

123

Mycopathologia (2013) 176:279–286

DOI 10.1007/s11046-013-9682-8

PS3, from the P. brasiliensis complex, and Paracoc-

cidioides lutzii—P. lutzii) [5–8]. S1 is the most widely

distributed species occurring in Brazil, Argentina,

Paraguay, Uruguay, Peru and Venezuela. PS2 occurs

in Venezuela and Brazil. PS3 is restricted to Colombia

and P. lutzii, occurs predominantly in the west central

region of Brazil [7, 8]. The PCM disease has two types

of manifestations: acute (juvenile) and chronic (adult).

The former accounts for less than 10 % of cases [2, 9,

10], affecting predominantly children and adolescents.

It can also affect individuals up to 35 years of age.

Gender distribution is similar, but there may be a slight

male predominance among young adults [10].

Involvement of the abdominal organs is most

frequent in children and young adults. The most

exuberant lesions are found in a form called lymph-

abdominal disease [4]. Autopsy studies have shown

that the liver is one of the organs most affected by

paracoccidioidomycosis [11–15].

Although liver involvement is more common in

children and young adults [10, 15–19], there is no

detailed study of liver involvement in the pediatric

population. Furthermore, PCM is not a disease of

compulsory notification, making the acquisition and

access to clinical and epidemiological data more

difficult. The present study aimed to describe clinical

and laboratory features of hepatic involvement by

paracoccidioidomycosis in pediatric patients.

Materials and Methods

This study comprised all patients under the age of

16 years diagnosed with PCM, managed at the UNI-

CAMP Teaching Hospital from January 1980 to

September 2007 (retrospective/96 patients) and new

cases from October 1, 2007 to April 31, 2010

(prospective/6 patients). Inclusion criterion consisted

of the diagnosis of paracoccidioidomycosis, estab-

lished by the detection of fungus in pathological

examination (lymph node, liver, bone, skin or bone

marrow) or cultures of pulmonary secretions or lymph

node fistulization. Exclusion criterion consisted of

cases associated with other diseases leading to liver

involvement or immunodeficiency, namely acquired

immunodeficiency syndrome caused by HIV. Cases of

HIV-related immunodeficiency were not included to

evaluate the clinical and laboratory features attribut-

able only to P. brasiliensis and not to the underlying

disease. We based the definition of hepatic involve-

ment on laboratory criteria because few patients

underwent liver biopsy (the gold standard for identi-

fication of liver involvement). Furthermore, the indi-

cation of liver biopsy occurred in a small number of

patients.

Hepatic involvement was defined as increased

levels of one or more liver enzymes (AST—aspartate

aminotransferase, ALT—alanine aminotransferase,

AF—alkaline phosphatase or GGT—gamma-glutam-

yltransferase). Elevation of alkaline phosphatase was

considered only if associated with an increased level

of any of the other three enzymes, due to the

possibility of bone involvement.

Enzyme level was considered elevated when

patient values were above the reference value used

by the Clinical Pathology laboratory of the Unicamp

Clinics Hospital at the time of blood sample collec-

tion. Hypoalbuminemia was defined as serum albumin

lower than 3.2 g/dL and hypergammaglobulinemia as

gamma globulin level higher than 2.1 g/dL.

We used the Gomez criteria to assess nutritional

status [20]:

• Normal Weight: Weight = 90–110 % of median

weight for chronological age

• Mild Malnutrition (grade I): Weight \90 % of

median weight for chronological age

• Moderate Malnutrition (grade II): Weight \75 %

of median weight for chronological age

• Severe Malnutrition (grade III): Weight \60 % of

median weight for chronological age

Concerning clinical and laboratory data at the time

of admission, this was a descriptive, analytical and

cross-sectional study. Follow-up data were also

obtained regarding clinical course and laboratory

tests. Registration of medical charts allowed for data

collection.

Logistic regression analysis with univariate and

stepwise multiple regression models for variable

criteria selection was used to analyze risk factors for

hepatic involvement. The receiver operating charac-

teristic curve (ROC curve) was used to obtain cutoff

values for baseline laboratory measurements to dif-

ferentiate between the presence and absence of liver

involvement. We used analysis of variance (ANOVA)

for repeated measures to compare numerical variables

related to laboratory tests between the two groups and

among the three sampling time periods (T0 at hospital

280 Mycopathologia (2013) 176:279–286

123

admission, T1 at 1 month of treatment and T6 at

6 months of treatment), followed by post hoc Turkey’s

multiple comparison to compare two groups at any

given time, and the contrast profile test to analyze

clinical course at the three time periods in each group.

The variables were transformed into ranks in the

absence of normal distribution. We used the Statistical

Package for Social Sciences (SPSS) for Windows

Institute Inc, 1989–1999, Chicago, IL, USA.), version

10.0.7, for database registration and preparation of

charts and graphs. The computer program Statistical

Analysis System (SAS) for Windows (Institute Inc,

2002–2003, Cary, NC, USA), version 9.1.3, was used

for statistical analysis. The significance level for

statistical tests was 5 % (p \ 0.05), and the confidence

interval was 95 % (when the confidence interval

contains the value one, the result is not statisti-

cally significant). The project was approved by

the Research Ethics Committee, under SISNEP

(CAAE-0493.0.146.000-07).

Results

From 1980 to 2010, 102 patients under age 16 were

diagnosed with paracoccidioidomycosis and received

follow-up care. Of these patients, 41 (40.1 %) had

liver involvement.

The main complaint was generalized lymph node

enlargement (25/41), followed by fever (6/41),

abdominal pain (3/41), increased abdominal size (2/

41), jaundice (2/41), asthenia (1/41), osteoarticular

pain (1/41) and diarrhea (1/41). Major clinical man-

ifestations were as follows: generalized lymphade-

nopathy (39/41 patients), weight loss (34/41) and fever

(32/41). Hepatomegaly was detected by physical

examination in 28/41 patients. The other clinical

manifestations observed were as follows: asthenia (26/

41), splenomegaly (24/41), pallor (23/41), anorexia

(19/41), skin and mucosal lesions (11/41), osteoartic-

ular symptoms (6/41), cough (3/41), edema (3/41),

lymph node fistulization (2/41) and localized lym-

phadenopathy (1/41). Approximately, one-third of the

patients (12/41) had jaundice. The only case of

bleeding disorder in this case series occurred in

patients with hepatic compromise. Ascites was infre-

quent (3/41). All patients with liver involvement had

an impaired lymphatic system (41/41). The spleen was

the second most affected organ (32/41). Other organs

or systems involved were as follows: the mucocuta-

neous tissue (13/41), osteoarticular tissue (11/41),

bone marrow (6/41), lungs (4/41), kidneys (2/41),

adrenals (1/41), tonsils (1/41), epiglottis (1/41), myo-

cardial infarction (1/41), pancreas (1/41) and testis (1/41).

Abdominal ultrasonography was performed in 34

patients: 27/34 had hepatomegaly, 23/34 had spleno-

megaly, 29/34 had adenopathy and one patient had a

normal examination. Eight patients with abdominal

adenopathy had jaundice, without bile duct dilatation.

Two deaths were recorded in patients with abdominal

adenopathy, and jaundice was associated in one of

these cases.

We present the frequency of altered levels of AST,

ALT, FALC, BD, albumin and gamma globulin in

patients with paracoccidioidomycosis and liver

involvement (Table 1).

Trimethoprim–sulfamethoxazole was the main drug

used for treatment, in both study groups, with or without

hepatic involvement. In the former group, a single-drug

treatment with trimethoprim–sulfamethoxazole was

given to 26/41 patients. An association with amphoter-

icin B, ketoconazole or itraconazole was required in

13/41 cases. The remaining patients (n = 2/41) received

a single-drug treatment with amphotericin B. Multiple-

drug treatment was more frequent in patients with liver

involvement (p\0.001).

We compared the clinical and laboratory charac-

teristics of patients with and without liver involvement

in Table 2.

Regarding the period from clinical progression to

death, univariate logistic regression analysis was

associated with liver involvement (OR = 22.52; CI

1.23–411.75) and albumin level at the time of

Table 1 Frequency of abnormal values of AST, ALT, FALC,

BD, albumin and gamma globulin in patients with paracoc-

cidioidomycosis and liver involvement

N %

Increased AST 12/33 36.4

Increased ALT 13/33 39.4

Increased FALC 25/33 75.7

Increased GGT 19/24 79.2

Increased BD 10/14 71.4

Hypoalbuminemia 36/39 92.3

Hypergammaglobulinemia 27/30 90.0

Mycopathologia (2013) 176:279–286 281

123

admission (OR = 0.154; CI 0.027–0.881). Multivar-

iate logistic regression analysis showed an association

with liver involvement; there was a 19.6-fold increase

in the risk of death. This analysis should be viewed

carefully because of the limited number of deaths in

this case series.

On univariate logistic regression analysis of liver

involvement, the variables malnutrition (p = 0.005),

age (p = 0.008), albumin (p \ 0.001) and hemoglobin

(p = 0.003) at baseline were statistically significant.

The results of multivariate logistic regression

analysis showed that baseline albumin levels, age

and malnutrition were associated with liver involve-

ment. Each 1-g/dL increase in baseline albumin levels

decreased the risk of liver involvement by 77.9 % or

each 1-g/dL decrease in baseline albumin levels

increased the risk of liver involvement by 4.5 times

(OR = 0.22; CI 0.085–0.571). Each 1-year increase in

age decreased the risk of liver involvement by 29.5 %

or any 1-year decrease in age increased the risk of liver

involvement by 41.8 % (OR = 0.75, CI 0.556–0.895).

Malnourished patients have a 5.5-fold higher risk of

liver involvement (OR = 5.51, CI 1.36–22.39).

It was possible to establish cutoff levels to deter-

mine the values of albumin, hemoglobin and lympho-

cyte with the highest probability of liver involvement

(albumin B3.05 g/dL, hemoglobin B9.2 g/dL and

lymphocytes B2,508/mm3).

The sensitivity of albumin for the diagnosis of

hepatic involvement is 82.1 %. The sensitivity of

Table 2 Comparison of clinical and laboratory variables between patients suffering from paracoccidioidomycosis, with and without

liver involvement, at time of admission

N Without liver

involvement

N With liver

involvement

p

Age (years) 61 9.5 ± 3.0 41 7.8 ± 2.5 0.009

Male gender 61 54 % 41 68.3 % 0.152

Time from onset of symptoms to diagnosis (days) 61 84.4 ± 82.3 41 82.9 ± 77.7 0.82

Malnutrition 59 52.5 % 41 80.5 % 0.008

Weight loss 61 60.6 % 41 82.9 % 0.029

Adynamia 61 36.0 % 41 63.4 % 0.012

Anorexia 61 18.6 % 41 46.3 % 0.0043

Pallor 61 11.5 % 41 56.1 % \0.001

Splenomegaly 61 3.3 % 41 58.5 % \0.001

Trimethoprim–sulfamethoxazole as single treatment 61 91.8 % 41 63.4 % \0.001

Deaths 61 0 41 6 0.002

Abdominal adenopathy 49 69.5 % 39 85 % 0.04

Hemoglobin g/dL 61 10.05 ± 1,9 41 8.8 ± 1.8 0.002

Leukocyte count/mm3 61 14194.4 ± 9378.9 41 14600 ± 8876.8 0.64

Lymphocyte count/mm3 61 2976.1 ± 1456.1 41 2399.5 ± 1764 0.003

Platelet count/mm3 60 405150 ± 178617 41 321951.2 ± 185441 0.023

Eosinophil count/mm3 61 2652 ± 4680 41 2427.5 ± 3256.6 0.78

Alpha-1 globulin g/dL 50 0.37 ± 0.11 30 0.45 ± 0.12 0.001

Alpha-2 globulin g/dL 49 0.96 ± 0.23 30 0.97 ± 0.29 0.91

Beta globulin g/dL 49 0.88 ± 0.21 30 0.79 ± 0.29 0.07

Albumin g/dL 61 3.1 ± 0.71 39 2.5 ± 0.6 \0.001

Gamma globulin g/dL 49 3.1 ± 1.12 30 3.4 ± 1.2 0.57

AST (U/L) 23 17.52 ± 10.34 33 47.6 ± 45.9 0.009

ALT (U/L) 23 14.83 ± 10.16 33 47.5 ± 53.8 0.003

AF (U/L) 19 440.74 ± 279.62 33 1119.7 ± 1168.3 0.019

GGT (U/L) 13 32.77 ± 36.32 24 262.9 ± 229.4 \0.001

BD (mg/dL) 5 0.14 14 2.9 ± 5.4 0.006

282 Mycopathologia (2013) 176:279–286

123

hemoglobin is 65.9 %, and lymphocytes have a

sensitivity of 70.7 %. The specificity for these vari-

ables is 59, 68.9 and 57.4 %, respectively. If albumin

is lower than 3.05 g/dL and hemoglobin is lower than

9.2 g/dL, the sensitivity for hepatic involvement is

61.54 % and specificity is 77.05 % (p \ 0.001).

We compared the laboratory measurements of

interest between groups with and without liver

involvement and among the three sampling time

periods (T0, T1 and T6). Only the data from patients

who had test samples in all three time periods were

computed.

The results were shown in Fig. 1. Briefly, hemo-

globin levels were significantly lower in patients with

liver involvement at all times. Hemoglobin had a trend

toward normalization in both groups, which is statis-

tically demonstrated at the time of admission. In

patients with liver involvement, this trend was signif-

icant for albumin and gamma globulin only at a later

time, after T1. Regarding leukocyte and eosinophil

counts, the downward trend can only be statistically

demonstrated in the group without liver involvement.

Concerning lymphocyte count, clinical course occurs

in opposite directions: a decline in the group without

liver involvement and a rise in patients with liver

involvement.

Discussion

Paracoccidioidomycosis was highly endemic in the

area where our patients acquired the disease [21, 31].

S1 and PS2 lineages of P. brasiliensis were isolated

from these areas [5, 6, 8]. Two major pediatric case

series originated in the state of Sao Paulo: the present

study and a study by Bellissimo-Rodrigues [21]. Only

few studies to date explore the clinical characteristics

of paracoccidioidomycosis in the pediatric population.

None of the previous pediatric case series [22–29],

acute/subacute cases [30] or those including all age

groups [15, 30–32] of paracoccidioidomycosis had a

larger number of pediatric patients than the two above-

mentioned case series.

Description of the frequency of hepatic involve-

ment in our study based on laboratory criteria is

relevant, in view of the size of this pediatric case

series. Furthermore, PCM is not a disease of compul-

sory notification in Brazil, hindering the acquisition of

clinical and epidemiological data.

The most common main complaint in our study was

lymphadenopathy, followed by fever and abdominal

pain. The same complaints were also observed by

other authors investigating PCM, PCM in children and

acute/subacute PCM [22, 24, 27, 28, 30]. The most

frequent clinical manifestations of this study are also

identified by other studies on paracoccidioidomycosis

in children and acute/subacute PCM: lymphadenopa-

thy, weight loss and fever [22, 24, 26–28, 30].

Abdominal ultrasonographic findings are similar to

other studies on acute PCM. Barbosa [22] and Ferreira

[30] described lymphadenopathy, hepatosplenomeg-

aly and one case of bile duct dilatation [22, 30]. A

frequent association between abdominal adenopathy

and hepatic involvement (p = 0.04) was expected.

Contamination in the liver with P. brasiliensis may

occur by lymphatic–hematogenous dissemination

from other tissues, particularly abdominal or by

contiguous invasion [4]. Although 8/12 icteric patients

had abdominal adenopathy, it is likely that lymph node

compression of major bile ducts was not the predom-

inant cause, since there was no report of bile duct

dilatation. Mechanisms responsible for jaundice in

PCM may be extrahepatic biliary tract obstruction

(choledocus infiltration by granulomas, extrinsic

compression by abdominal lymph nodes or choledo-

cus retraction due to fibrosis) or granulomatous

hepatitis. [15, 16]. Histopathological examination of

the liver shows a predominantly portal granulomatous

process, and interlobar bile duct compression with

cholestasis may be present in the periphery [4, 33].

Statistical differences were found in some clinical

variables evaluated in a comparison between patients

with and without liver involvement. Liver involve-

ment occurred in younger patients. The clinical

presentation was shown to be more exuberant in

patients with liver involvement, with a high percent-

age of malnourished individuals and increased rate of

complaints such as weight loss, malaise, anorexia and

pallor. However, this was not associated with a shorter

period between onset of complaints and diagnosis. A

common complaint of pallor can be correlated with a

lower mean hemoglobin level in patients with hepatic

involvement. A study showed that malnutrition is an

important factor contributing to death by paracoccid-

ioidomycosis [34].

Hepatomegaly is described in paracoccidioidomy-

cosis, ranging in frequency from 41.6 to 71.1 % in

both studies restricted to children/adolescents and also

Mycopathologia (2013) 176:279–286 283

123

studies evaluating the acute/subacute form of the

disease [16, 22, 24, 26, 28, 30]. The frequency of

hepatic involvement in this study is in agreement with

necropsy studies that are not restricted to the pediatric

population, whose proportion of liver injury ranges

from 21 to 56.7 % [4, 35].

Different results between groups at each time - T0, T1 and T6 (p= 0.002)

Different results over the period in both groups - T0≠T1, T0≠T6 (p=0.001)

T0 T1 T60

1

2

3

4

5

6Without liver involvement.With liver involvement.

Time

T0 T1 T61

2

3

4

5

Without liver involvementWith liver involvement.

Alb

umin

Time

T0 T1 T6

-2000

0

2000

4000

6000

8000

10000

Without liver involvement.With liver involvement.

Eos

inop

hils

Time

T0 T60

1000

2000

3000

4000

5000

6000

Without liver involvementWith liver involvement

Time

T0 T1 T60

5000

10000

15000

20000

25000

30000Without liver involvement.With liver involvement

Leuk

ocyt

es

Time

T0 T1 T66789

1011121314

Without liver involvement.With liver involvement

Hem

oglo

bin

Time

Lym

phoc

ytes

Gam

ma-

glob

ulin

Different results between groups at T6 and different results over the period in group without liver involvement ( ≠ ≠ and in group with liver involvement ≠ (p<0.001)

Different results between groups at T1 and T6 (p= 0.002)

Different results over the period only in group without liver involvement - T0≠T1, T0≠T6 (p<0.001)

Different results over the period only in group without liver involvement (T0≠T1, T0≠T6, T1≠T6); (p<0.001)

Different results over the period in group without liver involvement (T0≠T1, T0≠T6, T1≠T6); and in group with liver involvement ( ≠

≠ ). (p<0.001)

Different results between groups at T1 and T6 (p= 0.002)

Different results over the period in group without liver involvement (T0 ≠T1, T0≠T6, T1≠T6); and in group with liver involvement (T1≠T6). (p<0.001)

Fig. 1 Comparison and follow-up of hemoglobin, albumin and gamma globulin and leukocyte, lymphocyte and eosinophil counts in

patients with and without liver involvement at T0, T1 and T6

284 Mycopathologia (2013) 176:279–286

123

Similar to the study results found by other authors,

our study showed anemia, eosinophilia, hypoalbumi-

nemia and hypergammaglobulinemia [22, 27, 30, 35,

36]. Factors associated with anemia in PCM may be a

direct action of P. brasiliensis on the bone marrow,

malnutrition, loss of nutrients by intestinal involve-

ment and myelotoxicity related to amphotericin B use.

Some authors assume that the frequent association

between PCM and intestinal parasitic disease may

contribute to anemia [30]. A higher percentage of

malnourished patients with hepatic involvement,

showing statistical difference, may be related to more

severe anemia in this group. Other factors can explain

anemia in chronic liver disease, such as folate

deficiency, hypersplenism, hemodilution, hemolysis

and bleeding esophageal varices [37].

There was a predominant increase in canalicular

enzymes in patients with hepatic involvement. Similar

to other studies, aminotransferase elevations are milder

and less frequent. Increased levels of aminotransferases

may be due to the damaging effect of bile retention or

the direct action of the fungus or its products and

granuloma on the hepatocyte [30, 35, 36].

Lower albumin values in patients with hepatic

involvement did not arise simply because of the

presence of P. brasiliensis in those individuals.

Hypoalbuminemia may be indicative of liver dys-

function. Another possible cause and aggravating

factor of hypoalbuminemia is intestinal lymphangiec-

tasia as a complication of the disease [38].

A higher proportion of multiple-drug treatment in

patients with liver involvement may indicate greater

disease severity in these cases. Furthermore, all six

deaths occurred in patients with hepatic involvement.

Cutoff levels for simple laboratory tests such as

hemoglobin, albumin and lymphocyte count can help

in the diagnosis of hepatic involvement, even without

the performance of invasive tests (such as liver

biopsy). Young patients can receive proper clinical

care, since hepatic involvement is associated with a

greater risk of severe disease and a worse prognosis

(either by impairment of liver functions or as a marker

of disseminated disease).

As observed in other studies not limited to pediatric

patients, hepatic involvement associated with jaundice

may lead to a worse prognosis [15, 16]. The initial

follow-up of pediatric patients with paracoccidioido-

mycosis and liver involvement should be frequent

(weekly) because clinical and laboratory presentations

are more severe with an increased risk of death.

References

1. Coutinho ZF, Silva D, Lazera M, Petri V, Oliveira RM,

Sabroza PC, Wanke B. Paracoccidioidomycosis mortality in

Brazil (1980–1995). Cad Saude Publica. 2002;18:1441–54.

2. Moreira APV. Paracoccidioidomicose: historico, etiologia,

epidemiologia, patogenese, formas clınicas, diagnostico labo-

ratorial e antıgenos. Bol Epidemiol Paul. 2008;51:11–24.

3. Brummer E, Castaneda E, Restrepo A. Paracoccidioido-

mycosis: an update. Clin Microbiol Rev. 1993;6:89–117.

4. Del-Negro G, Lacaz CS, Fiorillo AM, editors. Paracoccid-

ioidomicose (Blastomicose sul-americana). Sarvier-Edusp:

Sao Paulo; 1982.

5. Matute DR, McEween JG, Montes BA, San-Blas G, Bagagli

E, Rauscher JT, Restrepo A, Morais F, Nino-Vega G, Taylor

JW. Cryptic speciation and recombination in the fungus

Paracoccidioides brasiliensis as revealed by gene geneal-

ogies. Mol Biol Evol. 2006;23:65–73.

6. Carrero LL, Nino-Vega G, Teixeira MM, Carvalho MJ,

Soares CM, Pereira M, Jesuino RS, McEwen JG, Mendoza

L, Taylor JW, Felipe MS, San-Blas G. New Paracoccidi-

oides brasiliensis isolate reveals unexpected genomic var-

iability in this human pathogen. Fungal Genet Biol.

2008;45:605–12.

7. Teixeira MM, Theodoro RC, Carvalho MJ, Fernandes L,

Paes HC, Hahn RC, Mendoza L, Bagagli E, San-Blas G,

Felipe MS. Phylogenetic analysis reveals a high level of

speciation in the Paracoccidioides genus. Mol Phylogenet

Evol. 2009;52:273–83.

8. Theodoro RC, Teixeira Mde M, Felipe MS, Paduan Kdos S,

Ribolla PM, San-Blas G, Bagagli E. Genus paracoccidioi-

des: species recognition and biogeographic aspects. PLoS

One. 2012;. doi:10.1371/journal.pone.0037694.

9. Franco M, Montenegro MR, Mendes RP, Marques SA,

Dillon ML, Mota NGS. Paracoccidioidomycosis: a recently

proposed classification of its clinical forms. Rev Soc Bras

Med Trop. 1987;20:129–32.

10. Shikanai-Yasuda MA, Telles Filho FQ, Mendes RP,

Colombo AL, Moretti ML. Guidelines in paracoccidioido-

mycosis. Rev Soc Bras Med Trop. 2006;39:297–310.

11. Pinto HB. La paracoccidiodosis brasiliensis como en-

fermedad sistemica. Comentarios a la casuistica venezol-

ana. Mycopathologia. 1961;15:90–114.

12. Pena CE. Deep mycotic infections in Colombia. A clini-

copathologic study of 162 cases. Am J Clin Path. 1967;

47:505–20.

13. Raphael A, Campana AO, Waimann J. Coma hepatico na bla-

stomicose sul-americana. Rev Ass Med Bras. 1964;10:151–4.

14. Teixeira F, Gayotto LC, Brito T. Morphological patterns of

the liver in South American blastomycosis. Histopathology.

1978;2:231–7.

15. Boccalandro I, Albuquerque FJM. Icterıcia e comprometi-

mento hepatico na blastomicose sul-americana. A proposito

de 10 casos. Rev Paul Med. 1960;56:350–66.

Mycopathologia (2013) 176:279–286 285

123

16. Daher RR, Vasconcelos WMP, Cardoso VM. Fıgado e a

blastomicose sul-americana. J Bras Med. 1973;25:83–90.

17. Marques SA. Paracoccidioimycosis: epidemiological, clin-

ical and treatment up-date. An Bras Dermatol. 2003;78:

135–50.

18. Campos MV, Penna GO, Castro CN, Moraes MA, Ferreira

MS, Santos JB. Paracoccidioidomycosis at Brasilias uni-

versity hospital. Rev Soc Bras Med Trop. 2008;41:169–72.

19. Paniago AM, Aguiar JI, Aguiar ES, da Cunha RV, Pereira

GR, Londero AT, Wanke B. Paracoccidioidomycosis: a

clinical and epidemiological study of 422 cases observed in

Mato Grosso do Sul. Rev Soc Bras Med Trop. 2003;

36:455–9.

20. Gomez F. Desnutricion. Bol Med Hosp Infant Mex.

1946;3:543–51.

21. Bellissimo-Rodrigues F, Machado AA, Martinez R. Para-

coccidioidomycosis epidemiological features of a 1,000-

cases series from a hyperendemic area on the southeast of

Brazil. Am J Trop Med Hyg. 2011;85:546–50.

22. Barbosa GL. Paracoccidioidomicose na crianca. Rev Pat

Trop. 1992;21:269–383.

23. Castro RM, Del Negro G. Particularidades clınicas da

paracoccidioidomicose na crianca. Rev Hosp Clin Fac Med

Sao Paulo. 1976;31:194–8.

24. Fonseca ERS, Pardal PPO, Severo LC. Paracoccidioidom-

icose em criancas em Belem do Para. Rev Soc Bras Med

Trop. 1999;32:31–3.

25. Pereira RM, Bucaretchi F, Barison Ede M, Hessel G, Tre-

soldi AT. Paracoccidioidomycosis in children: clinical

presentation, follow-up and outcome. Rev Inst Med Trop

Sao Paulo. 2004;46:127–31.

26. Londero AT, Goncalves AJR, Cruz MLS, Rozembaum R,

Cunha RQ, Machado ES, Viera ARM, Carvalho FG, Braga

MP, Azevedo ECL, Wanke B, Cruz MFF, Menezes JA.

Paracoccidioidomicose disseminada ‘‘infanto-juvenil’’ em

adolescentes. Relato de quatro casos e revisao da literatura.

Arq Bras Med. 1987;61:5–12.

27. Londero AT, Goncalves AJR, Terra GMF, Nogueira SA.

Paracoccidiodomycosis in Brazilian children. A critical

review (1911-1994). Arq Bras Med. 1996;70:197–203.

28. Nogueira MGS, Andrade GMQ, Tonelli E. Clinical evolu-

tion of paracoccidioidomycosis in 38 children and teenag-

ers. Mycopathologia. 2006;161:73–81.

29. Terra GMF, Rios-Goncalves AJ, Londero AT, Braga MP,

Ouricuri AL, Mesquita CC, Marino JCA, Ervilha LM,

Vieira ARM, Deker-Macher S, Duarte DMA. Paracoccidi-

oidomicose em criancas. Apresentacao de casos. Arq Bras

Med. 1991;65:8–15.

30. Ferreira MS. Contribuicao para o estudo clınico-laboratorial

e terapeutico da formacao juvenil da paracoccidioidomi-

cose. Rev Pat Trop. 1993;22:267–406.

31. Blotta MH, Mamoni RL, Oliveira SJ, Nouer SA, Papai-

ordanou PM, Goveia A, Camargo ZP. Endemic regions of

paracoccidioidomycosis in Brazil: a clinical and epidemi-

ologic study of 584 cases in the southeast region. Am J Trop

Med Hyg. 1999;61:390–4.

32. Valle ACP. Tratamento da paracoccidioidomicose: estudo

retrospectivo de 500 casos Analise clınica, laboratorial e

epidemiologica. An Bras Derm. 1992;67:251–4.

33. Brito T, Castro RM, Shiroma M. Biopsia hepatica na bla-

stomicose sul-americana. Rev Inst Med Trop Sao Paulo.

1968;10:188–91.

34. Santo AH. Tendencia da mortalidade relacionada a para-coccidioidomicose, Estado de Sao Paulo, Brasil, 1985 a

2005: estudo usando causas multiplas de morte. Rev Panam

Salud Publica. 2008;23:313–23.

35. Hildebrand TM, Rosario Filho NA, Telles Filho FQ, Costa

O, Miasaki N, Miyaki M. Paracoccidioidomicose na cri-

anca: aspectos clınicos-laboratoriais em 25 casos. J Ped.

1987;63:92–6.

36. Nogueira MGS, Andrade GMQ, Tonelli E, Diniz SN, Goes

AM, Cisalpino PS. Aspectos laboratoriais evolutivos de

criancas em tratamento da paracoccidioidomicose. Rev Soc

Bras Med Trop. 2006;39:478–83.

37. Gonzalez-Casas R, Jones EA, Moreno-Otero R. Spectrum

of anemia associated with chronic liver disease. World J

Gastroenterol. 2009;15:4653–8.

38. Troncon LE, Martinez R, Meneghelli UG, De Oliveira RB,

Iazigi N. Protein-losing enteropathy in paracoccidioidomyco-

sis. Rev Hosp Clin Fac Med Sao Paulo. 1981;36:172–8.

286 Mycopathologia (2013) 176:279–286

123