Vojnosanit Pregl 2016; 73(1): 21–25.

VOJNOSANITETSKI PREGLED Page 21

UDC: 612.57:616.9
ORIGINAL ARTICLE
DOI: 10.2298/VSP140820128P

Fever of unknown origin: Most frequent causes in adults patients

Febrilno stanje nejasne etiologije: najčešći uzroci kod odraslih bolesnika

Biljana Popovska-Jovičić, Predrag Čanović, Olgica Gajović, Ivana Raković,

Željko Mijailović

Clinic of Infectious Disease, Clinical Center of Kragujevac, Kragujevac, Serbia

Abstract Apstrakt

Background/Aim. The differential diagnosis of fever of un- Uvod/Cilj. Diferencijalna dijagnoza febrilnih stanja nepo-
known origin (FUO) includes more than 200 different diseases znatog porekla (fever of unknown origin – FUO), obuhvata
and conditions. The aim of this study was to identify the most preko 200 različitih bolesti i stanja. Cilj rada bio je da se
frequent causes of FUO among adult patients according to odrede najčešći uzroci FUO prema polu i životnom dobu.
gender and age. Methods. The study included 74 patients ex- Metode. Ispitivanjem su obuhvaćena 74 bolesnika koja su
amined from June 2010 to June 2013 at the Infectious Disease u periodu od juna 2010. do juna 2013. godine bila lečena u
Clinic, Clinical Center Kragujevac in Serbia, according to the Infektivnoj klinici Kliničkog centra u Kragujevcu prema
defined criteria for FUO. The patients were divided according definisanim kriterijumima za FUO. Prema dijagnozi boles-
to the diagnosis into four groups: infectious, malignant, rheu- nici su svrstani u četiri grupe: infektivne, maligne, reuma-
matic and “other diseases”. A cause of febricity could not be tološke i “druge bolesti”. Kod dela bolesnika nije utvrđen
estabilshed in a portion of subjects, and they comprised the uzrok febrilnosti i oni su činili grupu sa nedijagnostikova-
group of undiagnosed cases. Results. Infectious diseases were nim bolestima. Rezultati. U našem istraživanju dominirale
dominant in the study, followed by rheumatic diseases, which su infektivne bolesti, zatim reumatske, najviše zastupljene
were most frequently found in women and the elderly. The dis- kod ženskog pola i starije populacije. Bolesti koje su se iz-
eases recognised as the most common causes of febricity were dvojile kao najčešći uzroci febrilnosti bile su subakutni ti-
subacute thyroiditis, subacute endocarditis, Still’s disease, reoiditis, subakutni endokarditis, Stilova bolest, reumatska
rheumatic polymyalgia with or without temporal arteritis, and polimijalgija sa ili bez temporalnog arteritisa i citomegalo-
cytomegalovirus infection. In 44% of the patients, the final di- virusna infekcija. Kod 44% febrilnih bolesnika završnu di-
agnosis was composed of only six clinical entities. Conclusion. jagnozu činilo je svega šest kliničkih entiteta. Zaključak.
The importance of establishing the diagnosis of rheumatic dis- Posebno se ističe značaj utvrđivanja dijagnoze reumatskih
ease is especially emphasised, in line with other authors’ re- bolesti, čiji je broj u porastu, što je u skladu sa istraživa-
search indicating the number of these diseases is on the rise. njima drugih autora. Dijagnostički pristup FUO treba uvek
The diagnostic approach to FUO should always be directed to da bude usmeren prema poznatoj frekvenciji bolesti.
the known frequency of diseases.

Key words: Ključne reči:

fever of unknown origin; diagnosis; diagnosis, febrilnost nejasne etiologije; dijagnoza; dijagnoza,
differential; age groups; sex. diferencijalna; životno doba, grupe; pol.

Introduction tiation between standard FUO and three other forms of

febricity accompanied by neutropenia, nosocomial infections,
According to the traditionally accepted definition by Pe- and human immunodeficiency virus (HIV), and a shorter dura-
tersdorf and Beeson 1 in 1961, fever of unknown origin (FUO) tion of hospital examinations - three days instead of seven.
means a repeated occurrence of fever above 38.3°C in the co- FUO represents a great challenge to clinicians because its
urse of three weeks, the origin of which remains unresolved differential diagnosis is composed of more disorders than in
after three visits to the outpatient doctor or seven days of hos- any other medical condition, encompassing both very rare and
pital examination. Thirty years later, with the development of relatively common entities, classifiable into four groups: infec-
modern diagnostic methods, Durack and Street 2 suggested tious, malignant, rheumatic and “other” diseases 3–5. The pub-
two significant modifications to the initial definition: differen- lished papers indicate that the differential diagnosis of fever
Correspondence to: Biljana Popovska-Jovičić, Clinic of Infectious Disease, Clinical Center Kragujevac, Zmaj Jovina 30, 34 000
Kragujevac, Serbia. E-mail: [email protected]
Page 22 VOJNOSANITETSKI PREGLED Vol. 73, No. 1

includes more than 200 diseases and conditions, none of which Table 1
with an incidence higher than 5% 6. A significant portion of Causes of fever of unknown origin (FUO) in
patients who remain undiagnosed in spite of the rapid deve- total of 74 adults patients
lopment in diagnostic possibilities cannot be neglected, and Patients
FUO patients’ diagnoses
number
recently published studies suggest their number is increasing 7.
Infectious diseases 28 (37.8%)
Many researchers have investigated causes of fever in the Cytomegalovirus infection 5
elderly compared to the young adult population, and their results HIV infection 2
suggest that the etiologic factors differ in these populations 8. Epstein–Barr virus infection 1
The aim of this research was to determine the most staphylococcal sepsis 1
common causes of FUO in patients treated at the Infectious leptospirosis 1
Diseases Clinic of the Clinical Center Kragujevac, Serbia, to brucellosis 1
determine the distribution of various causes of FUO by gen- ITU 1
der, and to determine the most common causes of FUO in Localised Infections
subacute endocarditis 8
patients younger and older than 65 years.
acute cholecystitis 2
tubo-ovarian abscess 1
Methods perinephric abscess 1
pyonephrosis 1
The study included 74 adult hospitalized patients examined diverticulitis 1
at the Infectious Diseases Clinic in Clinical Center Kragujevac. iliopsoas abscesses 1
All the procedures were approved by the Ethical Committee of pulmonary abscess 1
Rheumatic diseases 19 (25.6%)
the Clinical Center in Kragujevac. Data were retrospectively
Still’s disease in adults 6
collected from the medical records in a three-years period, from systemic lupus erythematosus 3
June 2010 to June 2013. Only the patients with standard FUO, Polymyalgia rheumatica 4
body temperature higher than 38.3°C on several occasions du- temporal arteritis 3
ring a three-week period and the origin of fever remaining Reiter's syndrome 3
unknown during three visits to the doctor, or after three days of Malignant diseases 11 (14.8%)
hospital stay, were included. Patients with neutropenia, nosoco- colonic neoplasm 3
mial infections and FUO during HIV infection were excluded breast cancer with pleural metastases 1
renal cancer with hepatic and pulmonary
from the study. First, all the patients were classified into infecti- 1
metastases
ous and non-infectious disease groups. We then divided the pa- lung cancer with thoracic spine metastases 1
tients into the four groups according to the causative disease: in- prostatic neoplasm 2
fectious, malignancies, rheumatic, and group of “other” diseases pancreatic neoplasm 1
not belonging to any of the previously mentioned clinical en- hematologic diseases
tities. This group included granulomatous diseases, subacute acute myeloid leukaemia 1
thyroiditis, inflammatory diseases of the digestive tract as Hodgkin’s disease 1
well as drug-induced fever, factitious fever, and habitual Other diseases 13 (17.5%)
hyperthermia. The patients in whom the cause of febricity subacute thyroiditis 7
Crohn’s disease 1
remained unknown were classified into the group of undiag-
Löfgren syndrome 2
nosed FUO cases. Classification was then analyzed in relati- granulomatous hepatitis 1
on to patient gender and age. The age limit between the artificially induced fever 1
younger and older population was set to 65 years. drug-induced fever 1
The observed data are presented in tabular and graphi- Unknown cause of FUO 3 (4.1%)
cal form. The results obtained were analysed using the met-
hods of descriptive statistics.
Table 2
Visualizing procedures commonly used for the diagnosis of
Results fever of unknown origin
Diagnostic Imaging Number
In 28 (37.8%) patients of the 74 examined a disease of Chest X-ray 65
infectious origin was diagnosed, while in 43 (58.1%) patients Abdominal ultrasonography 68
a non-infectious cause of fever was found. In three (4.1%) Abdominal computed tomography (CT) 27
patients, the cause was not found (Table 1). Thorax computed thomography (CT) 16
Visualizine procedures commonly used for the diagno- Nuclear magnetic resonance (NMR) 7
sis of FUO are shown in Table 2. Positron emission tomography (PET) 2
In Table 1 it is shown that the majority of patients, 28
(37.8%), had infectious disease followed by 19 patients The diseases recognised as the most common causes of
(25.6%) with rheumatic disease, 13 patients (17.5%) had fever were subacute thyroiditis, subacute endocarditis, Still’s
“other disease”, and, finally, 11 patients (14.8%) had malig- disease, polymyalgia rheumatica and temporal arteritis, and
nant disease. The group V was composed of three (4.1%) pa- cytomegalovirus infection. The listed diseases were diagno-
tients with undiagnosed diseases. sed in as many as 33 (44%) of the total number of patients

Popovska-Jovičić B, et al. Vojnosanit Pregl 2016; 73(1): 21–25.

Vol. 73, No. 1 VOJNOSANITETSKI PREGLED Page 23

diagnosed with FUO, while the most frequent diagnoses Among the patients in our study 48 (64.8%) were under
were subacute thyroiditis and subacute endocarditis, together 65 years of age (Figure 2). The most common diagnosis was
are found to be the cause in every fifth patient with FUO. that of infectious disease, in 21 (43.7%) patients. Viral disea-
Three patients (4.1%) entered the group of unknown cau- ses dominated among them, and were found in eight patients.
ses of FUO. Following discharge from the Infectious Diseases Cytomegalovirus infection had the highest incidence, verifi-
Clinic, an undiagnosed patient experienced spontaneous subsi- ed by serological methods (ELISA test) in five patients. The
ding of febricity, while maintaining good general condition second most common diagnosis was rheumatic diseases, fo-
during the follow-up on the outpatient basis. Autopsy of the und in 10 patients (20.8%), especially Still’s disease,
second patient revealed malignancy, namely, pancreatic can- followed by systemic lupus erythematosus. Malignant disea-
cer. In the third male patient, three months of febricity accom- ses comprised a significant percentage of conditions in the
panied by biohumoral inflammatory syndrome were followed population under 65, occurring in nine (18.7%) patients.
by respiratory failure and death. The diagnosis remained Hematologic diseases were only found within this populati-
unknown despite exhaustive investigations. on, namely acute myeloid leukemia and Hodgkin’s disease.
The aim of the study was also to determine gender-wise Both patients were under 40 years of age.
distribution of different FUO factors. The study included 36 In the age group over 65 years, comprising 26 (35.1%)
(48.65 %) male and 38 (51.35 %) female patients. The obtained patients, the largest number of patients belonged to the group
results suggested that in the female population, the most com- with rheumatic conditions, nine (34.6%) patients. The group
mon were rheumatic diseases, 12 (31.5%). They are followed by was dominated by polymyalgia rheumatica and temporal ar-
infections in 11 (28.9%) of the patients, “other diseases” in 8 teritis as separate clinical entities. The group with infectious
(21.1%), and malignant diseases in 7 (18.4%) of the patients diseases was the second largest (Figure 2). Localised infecti-
(Figure 1). All FUO cases in the female patients were diagno- ons, especially subacute endocarditis, were verified in seven
sed. Most of the male patients, 17 (47.2%), had infectious disea- patients in this group. Subacute thyroiditis dominated the
se, followed by significantly lower percentages of rheumatic, 7 “other diseases” group, diagnosed using the radioactive iodi-
(19.4%), “other diseases”, 5 (13.8%), and malignancies, 4 ne fixation test. Malignancies were confirmed using diagnos-
(11.1%). Three (8.3%) of the patients entered the group of tic imaging and histopathologic findings in 2 (7.6%) of the
unknown FUO factors (Figure 1). patients.

Fig. 1 – Most common etiologies of fever of unknown origin according to gender.

Fig. 2 – Most common etiologies of fever of unknown origin (FUO) according to age
■ – patient with FUO under 65 years old; ■ – patients with FUO above 65 years old.

Popovska-Jovičić B, et al. Vojnosanit Pregl 2016; 73(1): 21–25.

Page 24 VOJNOSANITETSKI PREGLED Vol. 73, No. 1

Discussion 40 years of age, also represents a significant cause of FUO in

the middle-aged population.
The share of etiologies in the diagnosis of FUO has Among the subjects in the “other diseases” group, subacute
continuously changed since 1960. In the past two decades, cer- thyroiditis was undeniably the most common one. This endocri-
tain authors have suggested a decreased incidence of infectious nologic condition was also dominant among female subjects in
and malignant diseases, with the increase in the frequency of our sample. Subacute thyroiditis is a known cause of FUO, alt-
rheumatic diseases 9, 10. A recent research continues to pinpoint hough the diagnosis may prove more difficult to make in the ab-
infections as the leading causes of FUO, which is consistent sence of typical symptoms, such as frontal neck pain 14.
with our results. In a group of non-infectious FUO, the causes Some authors set hematological diseases and colorectal
according to other authors 11 are, malignancies, as the most cancer apart as the common causes of FUO in the group with
numerous, followed by rheumatic diseases. However, our re- malignancies 10. These diseases were also verified in our sam-
search indicates that the leading FUO causes among the non- ple, especially in the young adult population. Apart from those
infectious diseases were rheumatic conditions, while malig- conditions mentioned above, no other malignancies causing
nancies appeared as low as the fourth place. febricity were detected.
The development of modern diagnostic procedures According to many authors, the diagnosis of FUO differs
continuously modifies the spectrum of diseases with FUO. between older and younger populations 8, a presumption also
Introduction of diagnostic imaging, such as echosonography, confirmed by our study. In the elderly population, the most
computed tomography and magnetic resonance imaging, has numerous are rheumatic conditions, while the younger popula-
contributed to the rapid diagnosis of solid tumours and ab- tion is dominated by infections, especially viral ones.
scesses, which in turn influenced the number of infectious Polymyalgia rheumatica and temporal arteritis, both as joint
and malignant diseases in our FUO sample 12. Table 2 shows and separate clinical entities, are most commonly found in the
visualizing procedures commonly used in the diagnostic of elderly, which is a conclusion drawn by many other authors 15.
FUO patients. The invention and widespread use of serologi- For these diseases to appear among the common causes, con-
cal analysis enabled faster detection of many viral diseases, tributing factors are certainly the subacute course of disease
which reduced their frequency within FUO 11. Besides and non-specific symptoms.
serology, in the diagnostic of other infections and infectious In the group of rheumatic patients under the age of 60,
diseases we used microbiology tests, as well as polymerase Still’s disease is the most frequent clinical entity. This
chain reaction. multisystem disease can affect joints, skin, eyes, liver and
In our research, subacute endocarditis was the most spleen, but not all patients exhibit all symptoms 16. A review
common disease in the infectious diseases group and was of the available literature revealed that many authors empha-
especially dominant in the elderly population. Subacute en- sise Still’s disease as an important cause of FUO which is the
docarditis was present in 28% of patients with infectious di- fact that should be given particular consideration because the
seases, a significant percentage given the available diagnos- criteria for the diagnosis require the exclusion of other rheu-
tic procedures such as echocardiography. Diagnose determi- matic, malignant and infectious diseases 17.
nation was based on characteristic clinical findings and fin- Due to lack of specific tests in the diagnostics of Still's
dings of vegetations on the hart valvules. Using hemoculture disease and rheumatic polymyalgia, the diagnosis was estab-
Streptococcus viridans was isolated in three patients, lished using immunological, serological and microbiological
Staphylococcus aureus in one, while in other patiens no eti- analysis by which we excluded other diseases 18.
logic agent was isolated. It is believed that changes in the he-
art valves as well as the increased likelihood of sepsis make
Conclusion
the elderly population more vulnerable to the development of
endocarditis 13. Our study shows that only six causes account for 44%
The obtained results indicate that cytomegalovirus infec- of final diagnoses. The influence of rheumatic diseases is
tion is the most common etiologic cause of viral disease. Cli- especially emphasised, particularly in elderly persons, in line
nical diagnosis of cytomegalovirus infection can be difficult with other authors’ research indicating that the number of
due to a high prevalence of asymptomatic infections and vari- these diseases is on the rise. A diagnostic approach to fever
ous clinical presentations of the disease. Increased activity of of unknown disease should always be directed to the known
liver enzymes and splenomegaly, were the main parameters frequency of causing diseases. The first step is to rule out the
which directed us towards the diagnosis of cytomegalovirus common causes of fever of unknown disease and then review
infection. Seroprevalence, ranging from 50% to 80% before the less know causes.

Popovska-Jovičić B, et al. Vojnosanit Pregl 2016; 73(1): 21–25.

Vol. 73, No. 1 VOJNOSANITETSKI PREGLED Page 25

R E F E R E N C E S

1. Petersdorf RG, Beeson PB. Fever of unexplained origin: report on http://emedicine.medscape.com/article/217675-overview

100 cases. Medicine (Baltimore) 1961; 40: 1−30. [cited 2012 July 8].
2. Durack DT, Street AC. Fever of unknown origin--reexamined 12. Bleeker-Rovers CP, Vos FJ, de Kleijn EM, Mudde AH, Dofferhoff
and redefined. Curr Clin Top Infect Dis 1991; 11: 35−51. TS, Richter C, et al. A prospective multicenter study on fever of
3. Knockaert DC, Vanderschueren S, Blockmans D. Fever of unknown unknown origin: the yield of a structured diagnostic protocol.
origin in adults: 40 years on. J Intern Med 2003; 253(3): 263−75. Medicine (Baltimore) 2007; 86(1): 26−38.
4. Cunha BA. Fever of unknown origin: focused diagnostic ap- 13. Turkulov V, Brkić S, Sević S, Marić D, Tomić S. Fever of un-
proach based on clinical clues from the history, physical ex- known origin in elderly patients. Srp Arh Celok Lek 2011;
amination, and laboratory tests. Infect Dis Clin North Am 139(1−2): 64−8.
2007; 21(4): 1137−87. 14. Cunha BA, Thermidor M, Mohan S, Valsamis AS, Johnson DH,
5. Ergönül O, Willke A, Azap A, Tekeli E. Revised definition of Brook M, et al. Fever of unknown origin: Subacute thyroiditis
'fever of unknown origin': limitations and opportunities. J In- versus typhoid fever. Heart Lung 2005; 33(6): 417−21.
fect 2005; 50(1): 1−5. 15. Tal S, Guller V, Gurevich A, Levi S. Fever of unknown origin in
6. Finch RG, Davies GR, Čivljak R. Fever of unknown origin. In: the elderly. J Intern Med 2002; 252(4): 295−304.
Begovac J, Božnović D, Lisić M, editors. Infectology. 1st ed. Za- 16. Uppal SS, Al-Mutairi M, Hayat S, Abraham M, Malaviya A. Ten
greb: Profil International; 2006. p. 123−7. (Croatian) years of clinical experience with adult onset Still's disease: is the
7. Horowitz HW. Fever of unknown origin or fever of too many outcome improving. Clin Rheumatol 2007; 26(7): 1055−60.
origins. N Engl J Med 2013; 368(3): 197−9. 17. Kucukardali Y, Oncul O, Cavuslu S, Danaci M, Calangu S, Erdem
8. Gleckman RA, Esposito AL. Fever of unknown origin in older H, et al. The spectrum of diseases causing fever of unknown
patients. Geriatrics 1984; 39: 46−51. origin in Turkey: a multicenter study. Int J Infect Dis 2008;
9. Gaeta GB, Fusco FM, Nardiello S. Fever of unknown origin: a 12(1): 71−9.
systematic review of the literature for 1995-2004. Nucl Med 18. Cunha BA. Fever of unknown origin caused by adult juvenile
Commun 2006; 27(3): 205−11. rheumatoid arthritis: the diagnostic significance of double quo-
10. Zenone T. Fever of unknown origin in adults: evaluation of 144 tidian fevers and elevated serum ferritin levels. Heart Lung
cases in a non-university hospital. Scand J Infect Dis 2006; 2004; 33(6): 417−21.
38(8): 632−8. Received on August 20, 2014.
11. Chan-Tack KM, Cunha BA, Bartlett J, Sanders CV, Talavera F. Revised on October, 2014.
Fever unknown origin. Medscape from WebMD 2011 [serial Accepted on November 6, 2014.
on the Internet]. Available from: Online First November, 2015.

Popovska-Jovičić B, et al. Vojnosanit Pregl 2016; 73(1): 21–25.