- Email: [email protected]
Fever of unknown origin: What is remarkable in the elderly in a developing country?
Journal of Infection (2006) 52, 399–404
www.elsevierhealth.com/journals/jinf
Fever of unknown origin: What is remarkable in the elderly in a developing country? ¨ nal*, Mustafa Cankurtaran, Mustafa C ˙Ibrahim Koral O ¸akar, Meltem Halil, ¨ ¨ ¨ nal, ˘u, Omru ¨m Uzun, Serhat U Zekeriya Ulger, Burcu Balam Dog ˘ul Servet Arıog Department of Internal Medicine, Hacettepe University Medical School, 06100 Ankara, Turkey Accepted 24 August 2005 Available online 25 October 2005
KEYWORDS Fever of unknown origin (FUO); Elderly; Tuberculosis
Summary Objectives: To investigate fever of unknown origin (FUO) in 97 patients and compare geriatric and adult population. Methods: We investigated 97 (22 elderly) patients with FUO using the criteria of Petersdorf and Beeson [Medicine 40 (1961) 1] hospitalized between January 1990 and May 2005 at Hacettepe University Hospital. Results: Infectious diseases were the most common cause in the adult (33.3%) and the elderly (45.5%) patients both. Neoplasms were seen in 18.7; 4.5% and collagen vascular diseases were diagnosed in 9.3; 4.5% of the adults and the elderly respectively. Tuberculosis accounted for 60% of all the infectious causes and empirical anti-tuberculous treatment served as a diagnostic method in 43% of the cases with tuberculosis. Lymphadenopathy was more common among the adults with FUO. A diagnosis could be reached in all the elderly patients with a very high erythrocyte sedimentation rate (ESRO100 mm/h). At the end of the hospitalization, 14.7% (11/75) of the adult patients and 13.6% (3/22) of the elderly patients died. Conclusion: Geriatric patients with FUO usually have characteristics similar to the adult patients with respect to the hospitalization time, diagnosis, and inpatient mortality. Lymphoid organ hyperplasia might be expected less frequently and very high ESR might be a more reliable indicator of systemic disease in the elderly. Empirical anti-tuberculous treatment plays an important diagnostic role in the developing countries with a higher prevalence of tuberculosis. Q 2006 The British Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction
* Corresponding author. Tel.: C90 31230 51148; fax: C90 31230 52302. ¨ nal). E-mail address: [email protected] (I.K. O
Age-associated physiologic changes render geriatric patients more susceptible to miscellaneous disease processes. Atypical presentations are common and symptomatology is difficult to interpret. Fever of
0163-4453/$30.00 Q 2006 The British Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jinf.2005.08.021
¨ nal et al. I.K. O
400 unknown origin (FUO), one of the most historical and challenging entity to the clinicians, may require a rather specific approach in this patient population. FUO was first defined by Petersdorf and Beeson about 45 years ago as body temperature over 38.3 8C at least for three weeks, with no diagnosis reached after 1 week of inpatient investigation.1 With the advent of medical practice, the duration of investigation was later restricted to at least three hospital days or three outpatient visits.2 In the geriatric population, infection, especially tuberculosis, was reported to be the most common cause, followed by connective tissue diseases and malignancies.3–5 Diagnosis of a treatable disease could be made most of the time in this group of patients.5 On the other hand, we know that the spectrum of FUO can be influenced by geographic factors, developmental status of the country and age. Although many series have been reported since 1961, there are few data on the causes of FUO especially in the elderly regarding the developing countries. In this study, we describe the etiologies of FUO among the patients referred to the Hacettepe University Hospital for adults—a 800-bed tertiary care hospital located in Ankara, the capital city of Turkey—with an emphasis on the geriatric population.
Patients and methods In this study, we reviewed the hospital charts of 400 patients with fever who were referred to the internal medicine clinic from different parts of Turkey, mostly central Anatolia, and hospitalized in our wards between 1990 and 2005. Ninety-seven of them fulfilled the criteria for FUO.2 Medical history and findings in physical examination were extracted from the hospital charts. Laboratory studies included complete blood count, peripheral blood smear, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), biochemical tests (urea, creatinine, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, bilirubins, albumin and globulin), urinalysis, tuberculin skin test, three sets of blood cultures, culture for urine, and chest X-ray. According to the diagnostic clues provided by medical history, physical examination and preliminary laboratory data, further tests were performed. When a diagnosis could not be reached by non-invasive methods, invasive procedures such as biopsies from bone marrow, liver, enlarged lymph node, laparoscopy and laparotomy were performed. Mortality was calculated at the end of hospitalization.
Descriptive statistics were generated for all study variables, including mean and SD for continuous variables and relative frequencies for categorical variables. Relations between subgroups were analysed by using c2-test for categorical variables and by t-test for continuous variables. Two-sided values of P!0.05 were considered statistically significant. The statistical analyses were performed with SPSS 11.0.
Results A total of 97 patients with FUO were included in this study. Fifty-two patients (53.6%) were male, 45 (46.4%) were female. Twenty-two patients (22.7%) were older than 65 years. The duration of fever at the time of first admission to a health centre was more than one month in 12 (54.5%) elderly and 44 (58.7%) adult patients. Twelve (54.5%) elderly and 41 (54.7%) adult patients had been previously admitted to another healthcare facility for investigation of the cause of FUO. The mean duration of hospitalization was 26.81G20.65 days for the elderly patients and 22.08G13.62 days for the adult patients (PZ0.320). The signs, physical examination findings and laboratory data were similar between the elderly and the adult groups except for a more common occurrence of lymphadenopathy among the adults with FUO (PZ0.027, Table 1). The mean ESR was 72.7 mm/h in the adults and 75.3 mm/h in the elderly (PZ0.08). A diagnosis could not be reached in 8 (36.4%) elderly and 26 (34.7%) adult patients. Infections were the most common causes of fever in both groups, detected in 35 (36.1%) patients. The other causes were neoplasms in 15 (15.5%), collagen vascular diseases in 8 (8.2%) and miscellaneous diseases in 5 (5.2%) patients. There were no Table 1 Occurrence of some signs, physical examination findings and laboratory data in the elderly and the adult patients with fever of unknown origin Features
Adult, (%) nZ75
Elderly, (%) nZ22
P value
Weight loss Hepatomegaly Splenomegaly Lympadenopathy Anemia Leucocytosis Thrombocytosis Hypoalbuminemia
37(49.0%) 23(31.0%) 17(23.0%) 22(30.7%) 58(78.0%) 58(77.4%) 56(75.6%) 27(36.0%)
8(36.0%) 3(14.0%) 1(4.5 %) 1(4.5%) 16(73.3%) 17(77.3%) 17(78.6%) 8(36.3%)
0.407 0.190 0.065 0.027 0.948 1.000 0.921 0.986
Fever of unknown origin in the elderly Table 2
401
Comparison of diagnosis of fever of unknown origin between the elderly and the adult population
Causes of FUO
Adult (%)
Elderly (%)
P value
Infectious diseases Tuberculosis Pulmonary tuberculosis Tuberculous lympadenitis Miliary tuberculosis Tuberculous meningitis Tuberculous peritonitis Infective endocarditis Brucellosis Enteric fever Pneumonia Pyelonephritis Meningitis Neoplasms Non-hodgkin’s lymphoma Hodgkin’s disease Myelocytic leukaemia Chronic lymphocytic leukaemia Ipsid Colon carcinoma Malignancy of unknown primary Collagen-vascular diseases FMF SLE Temporal arteritis CNS vasculitis Still’s disease Seronegative spondyloarthropathy Reactive arthritis Miscellaneous diseases TTP Fenitoin hypersensitivity Subacute thyroiditis Contrast induced hyperthyroidism Sarcoidosis Undiagnosed
25(33.3%) 15 8 3 1 2 1 2 3 3 – 1 1 14(18.7 %) 6 2 2 1 1 1 1 7(9.3%) 2 1 – 1 1 1 1 3(4.0 %) 1 – 1 – 1 26(34.7 %)
10(45.5%) 6 4 1 1 – – 3 – – 1 – – 1(4.5%) 1 – – – – – – 1(4.5%) – – 1 – – – – 2(9.1%) – 1 –
0.298
– 8(36.4 %)
0.178
0.678
0.342
0.883
Ipsid, immunoproliferative small intestinal disease; FMF, familial mediterranean fever; SLE, systemic lupus erythematosus; CNS, central nervous system; TTP, thrombotic thrombocytopenic purpura.
differences between elderly and adult patients with regard to the cause of FUO (Table 2). Tuberculosis was the most common disease among all the infectious causes of FUO, and pulmonary tuberculosis was the most common type of involvement accounting for 57% (12/21) of all tuberculosis cases. Empirical anti-tuberculous therapy served as a diagnostic method in 43% (9/21) of the tuberculosis cases. A definitive diagnosis was reached in 12 patients (12.3%) with the aid of microbiological and serological tests, in 11 (11.3%) by imaging studies, and 28 (28.8%) by histopathological examination of the biopsy material (Table 3). Thoracic computerized tomography (CT) provided a diagnostic clue in 27% (20/73) of the
cases, and abdominal CT in 17% (12/68). The yield of the remaining methods was under %10 (Table 4). Biopsy was performed in 76% (74/97) of the cases. All of these tissue samples were obtained from the components of reticuloendothelial system (RES), except for the three from the temporal artery, the intestine and the kidney. Among the patients in whom a cause of FUO could be identified, biopsy was contributory in 35% (5/14) of the elderly patients and in 27% (14/52) of the adult patients (PZ0.75). At the end of the hospitalization, 14.7% (11/75) of the adult patients and 13.6% (3/22) of the elderly patients died. The two groups were not significantly different for the mortality (PZ1.00).
¨ nal et al. I.K. O
402 Table 3
The methods that led to a definitive diagnosis of the cause of fever of unknown origin
Diagnostic method that led to determination of the definite cause of FUO
No. of cases
Microbiology and serology Blood culture Examination of sputum for acid resistant bacillus of M. tuberculosis PCR for M. tuberculosis in sputum PCR for M. tuberculosis in bone marrow Salmonella serology Brucella agglutination Autoimmune serology for lupus CSF culture Imaging Transesophageal echocardiography Transthoracic echocardiography Abdomen CT Cranial MRI Biopsy Lymph node Inguinal Axillary Cervical Supraclavicular Liver (One during laparatomy) Bone marrow aspiration and biopsy Splenectomy (During laparotomy) Small intestine (By capsule endoscopy) Temporal artery Nefrectomy material Others Response to empirical anti-tuberculous treatment Thyroid function tests Medical history of phenytoin use Clinical response to plasmapheresis
12 3 1 2 1 2 1 1 1 11 4 5 1 1 28 8 3 2 2 1 7 7 3 1 1 1 12 9 1 1 1
Discussion In this study, we found that the diagnoses of FUO in the elderly did not differ from the adults. This is in contradiction with a series by Knockaert et al. implying that multisystem diseases (including connective tissue disorders and temporal arteritis) are more frequently associated with the etiology of Table 4
FUO in the elderly.6 On the other hand infections have been reported to be the most frequent cause by the results of several studies on the elderly and most of the studies on the general population compatible with the data from our series.3–5,7 Tuberculosis accounted for 60% of the infectious causes in our both geriatric and adult groups. The prevalence of tuberculosis within the infectious
Some frequently used methods that contributed to the diagnosis of fever of unknown origin
Diagnostic test
Number of patients who underwent the test
Diagnostic clue provided by the test, (%)
Thoracic CT Abdominal CT Chest X-ray Abdominal USG Bone marrow aspiration and biopsy Peripheric blood smear Brucella agglutination
73 68 97 62 74 89 86
20 (27) 12 (17) 7 (7) 4 (6) 4 (6) 4 (4) 2 (2)
CT, computerized tomography; USG, ultrasonography.
Fever of unknown origin in the elderly diseases was previously reported to be 37 and 70% in two series of FUO from Turkey, 8,9 24% in one series from India10 and 43.6% in one series from China.11 In reports from Europe, the proportion of tuberculosis has been 20–35%.3,12,13 A higher prevalence of tuberculosis as a cause of FUO in our elderly patients is not surprising since this disease is endemic in our country like the other developing countries. Diagnosis of tuberculosis is rather difficult in the geriatric population for several reasons. First, elderly patients are at increased risk of infection, particularly because of reactivation of early disease and the presentation may not be as distinct as in young people as far as symptomatology and positive purified protein derivative (ppd) test response are concerned.5 Second, a positive ppd test result is not very helpful in countries where a national vaccination program is carried out. In clinically highly probable cases where bacteriological or pathological confirmation is not possible, a course of empirical chemotherapy is safe and would prevent morbidity and mortality 14 In our series, empirical anti-tuberculous therapy played an important diagnostic as well as a therapeutic role in almost half of the cases. Diagnosis was made with the aid of response to empirical anti-tuberculous treatment in 43% (9/21) of the tuberculosis cases in whom all cultures and histopathological examinations were negative. Although the physiology of thermoregulation and pathogenesis of fever appears to be altered with advancing age, our geriatric patients had fever response similar to our adult patients as well as similar acute phase reactions including leucocytosis, thrombocytosis, high ESR and hypoalbuminemia. On the other hand, we observed hepatomegaly, splenomegaly and lymphadenopathy less frequently in the elderly population. Alterations to the immune system in the elderly are generally viewed as a deterioration of immunity, leading to the use of the term immunosenescence, a complex subject best defined as a decline in cell-mediated immunity, particularly with respect to T-cell function.15 Therefore, the absence of clinically apparent reticuloendothelial system hyperplasia as organomegaly and lympadenopathy should not make the possibility of a systemic disease unlikely in geriatric population. Interestingly, a diagnosis could be made significantly more commonly in patients with ESR R100 mm/h (PZ0.043). The proportion to reach a diagnosis within the high sedimentation rate group was %100 in the elderly and %77 in the adult patients. Regarding the elderly patients with FUO high sedimentation rate might be a more important indicator for the presence of a systemic disease and
403 an agressive diagnostic approach might be more rational in these patients. We could reach a diagnosis in 63.6% of our geriatric cases and 65.3% of our adult cases. This is in accordance with the undiagnosed case ratio of 7– 38% obtained in different studies in developing and developed countries both.10,16–21 It is known that patients with unexplained FUO generally have a good prognosis,22 however, there are not much data specifically on the elderly. Unfortunately, we did not have a long-term follow-up of these geriatric patients with unexplained FUO. Abdominal CT provided a diagnostic clue of 17%, similar to that reported by Quinn et al. (19%).23 Although the diagnostic yield of thoracic CT lagged behind those of abdominal CT and USG in various FUO series,24,25 it provided the highest percentage of any diagnostic clue among our patients. This may be caused by the fact that tuberculosis is more prevalent in our country like the other developing countries and thoracic CT provided some basis for empirical (fibrous sequela in the apices, calcified lymph nodes, etc.) or definitive anti-tuberculous treatment. Invasive methods helped us to establish a diagnosis in 19.6% of our cases with no difference between the elderly and the adult population. It was 43–48% in previously reported three series.8,9,18 The difference may be attributed to the weakness of all four series including ours as far as the number of patients are considered. It may also result from the types of the invasive procedures and their diagnostic value, which show extensive variability among the studies. For example, the contribution of laparotomy to the diagnosis of FUO was reported as 27–100% in several FUO series.26 The improvement in imaging techniques and the difference in the experience and technology of a particular radiology department may diminish the need to perform an invasive procedure that is likely to provide a diagnosis. In our series, most of our patients underwent bone marrow biopsy and the diagnostic yield was under 10%, confirming that such invasive methods should not be preferred without any clinical, laboratory or radiological clue. The only biopsy that may be often rewarding in the absence of prior localizing information is temporal artery biopsy in elderly patients with a very high ESR.6 In conclusion, geriatric patients with FUO usually have characteristics similar to the adult patients with respect to the hospitalization time, diagnosis, and inpatient mortality. However, lymphoid organ hyperplasia might be expected less frequently. Very high ESR might be a more reliable indicator of systemic disease in the elderly. Finally, empirical
¨ nal et al. I.K. O
404 anti-tuberculous treatment plays an important diagnostic role in developing countries with a higher prevalence of tuberculosis.
14. 15.
References 1. Petersdorf RG, Beeson PB. Fever of unknown origin: report on 100 cases. Medicine 1961;40:1–30. 2. Durack DT, Street AC. Fever of unknown origin—reexamined and redefined. Curr Clin Top Infect Dis 1991;11:35–51. 3. Esposito AL, Gleckman RA. Fever of unknown origin in the elderly. J Am Geriatr Soc 1978;26(11):498–505. 4. Norman DC, Yoshikawa TT. Fever in the elderly. Infect Dis Clin North Am 1996;10(1):93–9. 5. Tal S, Guller V, Gurevich A, Levi S. Fever of unknown origin in the elderly. J Intern Med 2002;252(4):295–304. 6. Knockaert DC, Vanneste LJ, Bobbaers HJ. Fever of unknown origin in elderly patients. J Am Geriatr Soc 1993;41(11): 1187–92. 7. Arnow PM, Flaherty JP. Fever of unknown origin. Lancet 1997;350(9077):575–80. 8. Tabak F, Mert A, Celik AD, Ozaras R, Altiparmak MR, Ozturk R, et al. Fever of unknown origin in Turkey. Infection 2003;31(6):417–20. 9. Saltoglu N, Tasova Y, Midikli D, Aksu HS, Sanli A, Dundar IH. Fever of unknown origin in Turkey: Evaluation of 87 cases during a nine-year-period of study. J Infect 2004;48(1):81–5. 10. Kejariwal D, Sarkar N, Chakraborti SK, Agarwal V, Roy S. Pyrexia of unknown origin: A prospective study of 100 cases. J Postgrad Med 2001;47(2):104–7. 11. Ma XJ, Wang AX, Deng GH, Sheng RY. A clinical review of 449 cases with fever of unknown origin. Zhonghua Nei Ke Za Zhi 2004;43(9):682–5. 12. Barrier J, Schneebeli S, Peltier P. Les fievres prolongees inexpliquees chez les personnes, agees. Concours Med 1982; 104:4679–89. 13. Ramos Rincon JM, Ramos Guevara R, Herrero Huerta F. Fever of unknown origin in internal medicine. The experience of
16.
17.
18.
19.
20.
21. 22.
23.
24.
25. 26.
Spanish authors over 20 years. An Med Interna 1997;14(11): 585–92. ´smail Y. Tuberculosis—are we missing the diagnosis? Y Singapore Med J 2002;43(4):172–6. Boren E, Gershwin ME. Inflamm-aging: Autoimmunity, and the immune-risk phenotype. Autoimmun Rev 2004;3(5): 401–6. Zhiyong Z, Bingjun L, Xiaoju L, Xinjian F, Ping F, Wenya W. Fever of unknown origin: A report from China of 208 cases. Int J Clin Pract 2003;57(7):592–6. de Kleijn EM, van Lier HJ, van der Meer JW. Fever of unknown origin (FUO): II. Diagnostic procedures in a prospective multicenter study of 167 patients.The Netherlands FUO Study Group. Medicine (Baltimore) 1997;76(6): 401–14. de Kleijn EM, van der Meer JW. Fever of unknown origin (FUO): Report on 53 patients in a Dutch university hospital. Neth J Med 1995;47(2):54–60. Deng GH, Wang AX. Clinical analysis of 130 patients with fever of unknown origin. Zhonghua Nei Ke Za Zhi 1991;30(3): 157–9 [also see p. 188–189]. Knockaert DC, Vanneste LJ, Vanneste SB, Bobbaers HJ. Fever of unknown origin in the 1980s. An update of the diagnostic spectrum. Arch Intern Med 1992;152(1):51–5. Norman DC. Fever in the elderly. Clin Infect Dis 2000;31: 148–51. Knockaert DC, Dujardin KS, Bobbaers HJ. Long term followup of patients with undiagnosed fever of unknown origin. Arch Intern Med 1996;156(6):618–20. Quinn MJ, Sheedy 2nd PF, Stephens DH, Hattery RR. Computed tomography of the abdomen in evaluation of patients with fever of unknown origin. Radiology 1980; 136(2):407–11. Mourad O, Palda V, Detsky AS. A comprehensive evidencebased approach to fever of unknown origin. Arch Intern Med 2003;163(5):545–51. Arnow PM, Flaherty JP. Fever of unknown origin. Lancet 1997;350(9077):575–80. Ozaras R, Celik AD, Zengin K, Mert A, Ozturk KR, Cicek Y, et al. Is laparotomy necesssary in the diagnosis of fever of unknown origin? Acta Chir Belg 2005;105(1):89–92.
www.elsevierhealth.com/journals/jinf
Fever of unknown origin: What is remarkable in the elderly in a developing country? ¨ nal*, Mustafa Cankurtaran, Mustafa C ˙Ibrahim Koral O ¸akar, Meltem Halil, ¨ ¨ ¨ nal, ˘u, Omru ¨m Uzun, Serhat U Zekeriya Ulger, Burcu Balam Dog ˘ul Servet Arıog Department of Internal Medicine, Hacettepe University Medical School, 06100 Ankara, Turkey Accepted 24 August 2005 Available online 25 October 2005
KEYWORDS Fever of unknown origin (FUO); Elderly; Tuberculosis
Summary Objectives: To investigate fever of unknown origin (FUO) in 97 patients and compare geriatric and adult population. Methods: We investigated 97 (22 elderly) patients with FUO using the criteria of Petersdorf and Beeson [Medicine 40 (1961) 1] hospitalized between January 1990 and May 2005 at Hacettepe University Hospital. Results: Infectious diseases were the most common cause in the adult (33.3%) and the elderly (45.5%) patients both. Neoplasms were seen in 18.7; 4.5% and collagen vascular diseases were diagnosed in 9.3; 4.5% of the adults and the elderly respectively. Tuberculosis accounted for 60% of all the infectious causes and empirical anti-tuberculous treatment served as a diagnostic method in 43% of the cases with tuberculosis. Lymphadenopathy was more common among the adults with FUO. A diagnosis could be reached in all the elderly patients with a very high erythrocyte sedimentation rate (ESRO100 mm/h). At the end of the hospitalization, 14.7% (11/75) of the adult patients and 13.6% (3/22) of the elderly patients died. Conclusion: Geriatric patients with FUO usually have characteristics similar to the adult patients with respect to the hospitalization time, diagnosis, and inpatient mortality. Lymphoid organ hyperplasia might be expected less frequently and very high ESR might be a more reliable indicator of systemic disease in the elderly. Empirical anti-tuberculous treatment plays an important diagnostic role in the developing countries with a higher prevalence of tuberculosis. Q 2006 The British Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction
* Corresponding author. Tel.: C90 31230 51148; fax: C90 31230 52302. ¨ nal). E-mail address: [email protected] (I.K. O
Age-associated physiologic changes render geriatric patients more susceptible to miscellaneous disease processes. Atypical presentations are common and symptomatology is difficult to interpret. Fever of
0163-4453/$30.00 Q 2006 The British Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jinf.2005.08.021
¨ nal et al. I.K. O
400 unknown origin (FUO), one of the most historical and challenging entity to the clinicians, may require a rather specific approach in this patient population. FUO was first defined by Petersdorf and Beeson about 45 years ago as body temperature over 38.3 8C at least for three weeks, with no diagnosis reached after 1 week of inpatient investigation.1 With the advent of medical practice, the duration of investigation was later restricted to at least three hospital days or three outpatient visits.2 In the geriatric population, infection, especially tuberculosis, was reported to be the most common cause, followed by connective tissue diseases and malignancies.3–5 Diagnosis of a treatable disease could be made most of the time in this group of patients.5 On the other hand, we know that the spectrum of FUO can be influenced by geographic factors, developmental status of the country and age. Although many series have been reported since 1961, there are few data on the causes of FUO especially in the elderly regarding the developing countries. In this study, we describe the etiologies of FUO among the patients referred to the Hacettepe University Hospital for adults—a 800-bed tertiary care hospital located in Ankara, the capital city of Turkey—with an emphasis on the geriatric population.
Patients and methods In this study, we reviewed the hospital charts of 400 patients with fever who were referred to the internal medicine clinic from different parts of Turkey, mostly central Anatolia, and hospitalized in our wards between 1990 and 2005. Ninety-seven of them fulfilled the criteria for FUO.2 Medical history and findings in physical examination were extracted from the hospital charts. Laboratory studies included complete blood count, peripheral blood smear, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), biochemical tests (urea, creatinine, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, bilirubins, albumin and globulin), urinalysis, tuberculin skin test, three sets of blood cultures, culture for urine, and chest X-ray. According to the diagnostic clues provided by medical history, physical examination and preliminary laboratory data, further tests were performed. When a diagnosis could not be reached by non-invasive methods, invasive procedures such as biopsies from bone marrow, liver, enlarged lymph node, laparoscopy and laparotomy were performed. Mortality was calculated at the end of hospitalization.
Descriptive statistics were generated for all study variables, including mean and SD for continuous variables and relative frequencies for categorical variables. Relations between subgroups were analysed by using c2-test for categorical variables and by t-test for continuous variables. Two-sided values of P!0.05 were considered statistically significant. The statistical analyses were performed with SPSS 11.0.
Results A total of 97 patients with FUO were included in this study. Fifty-two patients (53.6%) were male, 45 (46.4%) were female. Twenty-two patients (22.7%) were older than 65 years. The duration of fever at the time of first admission to a health centre was more than one month in 12 (54.5%) elderly and 44 (58.7%) adult patients. Twelve (54.5%) elderly and 41 (54.7%) adult patients had been previously admitted to another healthcare facility for investigation of the cause of FUO. The mean duration of hospitalization was 26.81G20.65 days for the elderly patients and 22.08G13.62 days for the adult patients (PZ0.320). The signs, physical examination findings and laboratory data were similar between the elderly and the adult groups except for a more common occurrence of lymphadenopathy among the adults with FUO (PZ0.027, Table 1). The mean ESR was 72.7 mm/h in the adults and 75.3 mm/h in the elderly (PZ0.08). A diagnosis could not be reached in 8 (36.4%) elderly and 26 (34.7%) adult patients. Infections were the most common causes of fever in both groups, detected in 35 (36.1%) patients. The other causes were neoplasms in 15 (15.5%), collagen vascular diseases in 8 (8.2%) and miscellaneous diseases in 5 (5.2%) patients. There were no Table 1 Occurrence of some signs, physical examination findings and laboratory data in the elderly and the adult patients with fever of unknown origin Features
Adult, (%) nZ75
Elderly, (%) nZ22
P value
Weight loss Hepatomegaly Splenomegaly Lympadenopathy Anemia Leucocytosis Thrombocytosis Hypoalbuminemia
37(49.0%) 23(31.0%) 17(23.0%) 22(30.7%) 58(78.0%) 58(77.4%) 56(75.6%) 27(36.0%)
8(36.0%) 3(14.0%) 1(4.5 %) 1(4.5%) 16(73.3%) 17(77.3%) 17(78.6%) 8(36.3%)
0.407 0.190 0.065 0.027 0.948 1.000 0.921 0.986
Fever of unknown origin in the elderly Table 2
401
Comparison of diagnosis of fever of unknown origin between the elderly and the adult population
Causes of FUO
Adult (%)
Elderly (%)
P value
Infectious diseases Tuberculosis Pulmonary tuberculosis Tuberculous lympadenitis Miliary tuberculosis Tuberculous meningitis Tuberculous peritonitis Infective endocarditis Brucellosis Enteric fever Pneumonia Pyelonephritis Meningitis Neoplasms Non-hodgkin’s lymphoma Hodgkin’s disease Myelocytic leukaemia Chronic lymphocytic leukaemia Ipsid Colon carcinoma Malignancy of unknown primary Collagen-vascular diseases FMF SLE Temporal arteritis CNS vasculitis Still’s disease Seronegative spondyloarthropathy Reactive arthritis Miscellaneous diseases TTP Fenitoin hypersensitivity Subacute thyroiditis Contrast induced hyperthyroidism Sarcoidosis Undiagnosed
25(33.3%) 15 8 3 1 2 1 2 3 3 – 1 1 14(18.7 %) 6 2 2 1 1 1 1 7(9.3%) 2 1 – 1 1 1 1 3(4.0 %) 1 – 1 – 1 26(34.7 %)
10(45.5%) 6 4 1 1 – – 3 – – 1 – – 1(4.5%) 1 – – – – – – 1(4.5%) – – 1 – – – – 2(9.1%) – 1 –
0.298
– 8(36.4 %)
0.178
0.678
0.342
0.883
Ipsid, immunoproliferative small intestinal disease; FMF, familial mediterranean fever; SLE, systemic lupus erythematosus; CNS, central nervous system; TTP, thrombotic thrombocytopenic purpura.
differences between elderly and adult patients with regard to the cause of FUO (Table 2). Tuberculosis was the most common disease among all the infectious causes of FUO, and pulmonary tuberculosis was the most common type of involvement accounting for 57% (12/21) of all tuberculosis cases. Empirical anti-tuberculous therapy served as a diagnostic method in 43% (9/21) of the tuberculosis cases. A definitive diagnosis was reached in 12 patients (12.3%) with the aid of microbiological and serological tests, in 11 (11.3%) by imaging studies, and 28 (28.8%) by histopathological examination of the biopsy material (Table 3). Thoracic computerized tomography (CT) provided a diagnostic clue in 27% (20/73) of the
cases, and abdominal CT in 17% (12/68). The yield of the remaining methods was under %10 (Table 4). Biopsy was performed in 76% (74/97) of the cases. All of these tissue samples were obtained from the components of reticuloendothelial system (RES), except for the three from the temporal artery, the intestine and the kidney. Among the patients in whom a cause of FUO could be identified, biopsy was contributory in 35% (5/14) of the elderly patients and in 27% (14/52) of the adult patients (PZ0.75). At the end of the hospitalization, 14.7% (11/75) of the adult patients and 13.6% (3/22) of the elderly patients died. The two groups were not significantly different for the mortality (PZ1.00).
¨ nal et al. I.K. O
402 Table 3
The methods that led to a definitive diagnosis of the cause of fever of unknown origin
Diagnostic method that led to determination of the definite cause of FUO
No. of cases
Microbiology and serology Blood culture Examination of sputum for acid resistant bacillus of M. tuberculosis PCR for M. tuberculosis in sputum PCR for M. tuberculosis in bone marrow Salmonella serology Brucella agglutination Autoimmune serology for lupus CSF culture Imaging Transesophageal echocardiography Transthoracic echocardiography Abdomen CT Cranial MRI Biopsy Lymph node Inguinal Axillary Cervical Supraclavicular Liver (One during laparatomy) Bone marrow aspiration and biopsy Splenectomy (During laparotomy) Small intestine (By capsule endoscopy) Temporal artery Nefrectomy material Others Response to empirical anti-tuberculous treatment Thyroid function tests Medical history of phenytoin use Clinical response to plasmapheresis
12 3 1 2 1 2 1 1 1 11 4 5 1 1 28 8 3 2 2 1 7 7 3 1 1 1 12 9 1 1 1
Discussion In this study, we found that the diagnoses of FUO in the elderly did not differ from the adults. This is in contradiction with a series by Knockaert et al. implying that multisystem diseases (including connective tissue disorders and temporal arteritis) are more frequently associated with the etiology of Table 4
FUO in the elderly.6 On the other hand infections have been reported to be the most frequent cause by the results of several studies on the elderly and most of the studies on the general population compatible with the data from our series.3–5,7 Tuberculosis accounted for 60% of the infectious causes in our both geriatric and adult groups. The prevalence of tuberculosis within the infectious
Some frequently used methods that contributed to the diagnosis of fever of unknown origin
Diagnostic test
Number of patients who underwent the test
Diagnostic clue provided by the test, (%)
Thoracic CT Abdominal CT Chest X-ray Abdominal USG Bone marrow aspiration and biopsy Peripheric blood smear Brucella agglutination
73 68 97 62 74 89 86
20 (27) 12 (17) 7 (7) 4 (6) 4 (6) 4 (4) 2 (2)
CT, computerized tomography; USG, ultrasonography.
Fever of unknown origin in the elderly diseases was previously reported to be 37 and 70% in two series of FUO from Turkey, 8,9 24% in one series from India10 and 43.6% in one series from China.11 In reports from Europe, the proportion of tuberculosis has been 20–35%.3,12,13 A higher prevalence of tuberculosis as a cause of FUO in our elderly patients is not surprising since this disease is endemic in our country like the other developing countries. Diagnosis of tuberculosis is rather difficult in the geriatric population for several reasons. First, elderly patients are at increased risk of infection, particularly because of reactivation of early disease and the presentation may not be as distinct as in young people as far as symptomatology and positive purified protein derivative (ppd) test response are concerned.5 Second, a positive ppd test result is not very helpful in countries where a national vaccination program is carried out. In clinically highly probable cases where bacteriological or pathological confirmation is not possible, a course of empirical chemotherapy is safe and would prevent morbidity and mortality 14 In our series, empirical anti-tuberculous therapy played an important diagnostic as well as a therapeutic role in almost half of the cases. Diagnosis was made with the aid of response to empirical anti-tuberculous treatment in 43% (9/21) of the tuberculosis cases in whom all cultures and histopathological examinations were negative. Although the physiology of thermoregulation and pathogenesis of fever appears to be altered with advancing age, our geriatric patients had fever response similar to our adult patients as well as similar acute phase reactions including leucocytosis, thrombocytosis, high ESR and hypoalbuminemia. On the other hand, we observed hepatomegaly, splenomegaly and lymphadenopathy less frequently in the elderly population. Alterations to the immune system in the elderly are generally viewed as a deterioration of immunity, leading to the use of the term immunosenescence, a complex subject best defined as a decline in cell-mediated immunity, particularly with respect to T-cell function.15 Therefore, the absence of clinically apparent reticuloendothelial system hyperplasia as organomegaly and lympadenopathy should not make the possibility of a systemic disease unlikely in geriatric population. Interestingly, a diagnosis could be made significantly more commonly in patients with ESR R100 mm/h (PZ0.043). The proportion to reach a diagnosis within the high sedimentation rate group was %100 in the elderly and %77 in the adult patients. Regarding the elderly patients with FUO high sedimentation rate might be a more important indicator for the presence of a systemic disease and
403 an agressive diagnostic approach might be more rational in these patients. We could reach a diagnosis in 63.6% of our geriatric cases and 65.3% of our adult cases. This is in accordance with the undiagnosed case ratio of 7– 38% obtained in different studies in developing and developed countries both.10,16–21 It is known that patients with unexplained FUO generally have a good prognosis,22 however, there are not much data specifically on the elderly. Unfortunately, we did not have a long-term follow-up of these geriatric patients with unexplained FUO. Abdominal CT provided a diagnostic clue of 17%, similar to that reported by Quinn et al. (19%).23 Although the diagnostic yield of thoracic CT lagged behind those of abdominal CT and USG in various FUO series,24,25 it provided the highest percentage of any diagnostic clue among our patients. This may be caused by the fact that tuberculosis is more prevalent in our country like the other developing countries and thoracic CT provided some basis for empirical (fibrous sequela in the apices, calcified lymph nodes, etc.) or definitive anti-tuberculous treatment. Invasive methods helped us to establish a diagnosis in 19.6% of our cases with no difference between the elderly and the adult population. It was 43–48% in previously reported three series.8,9,18 The difference may be attributed to the weakness of all four series including ours as far as the number of patients are considered. It may also result from the types of the invasive procedures and their diagnostic value, which show extensive variability among the studies. For example, the contribution of laparotomy to the diagnosis of FUO was reported as 27–100% in several FUO series.26 The improvement in imaging techniques and the difference in the experience and technology of a particular radiology department may diminish the need to perform an invasive procedure that is likely to provide a diagnosis. In our series, most of our patients underwent bone marrow biopsy and the diagnostic yield was under 10%, confirming that such invasive methods should not be preferred without any clinical, laboratory or radiological clue. The only biopsy that may be often rewarding in the absence of prior localizing information is temporal artery biopsy in elderly patients with a very high ESR.6 In conclusion, geriatric patients with FUO usually have characteristics similar to the adult patients with respect to the hospitalization time, diagnosis, and inpatient mortality. However, lymphoid organ hyperplasia might be expected less frequently. Very high ESR might be a more reliable indicator of systemic disease in the elderly. Finally, empirical
¨ nal et al. I.K. O
404 anti-tuberculous treatment plays an important diagnostic role in developing countries with a higher prevalence of tuberculosis.
14. 15.
References 1. Petersdorf RG, Beeson PB. Fever of unknown origin: report on 100 cases. Medicine 1961;40:1–30. 2. Durack DT, Street AC. Fever of unknown origin—reexamined and redefined. Curr Clin Top Infect Dis 1991;11:35–51. 3. Esposito AL, Gleckman RA. Fever of unknown origin in the elderly. J Am Geriatr Soc 1978;26(11):498–505. 4. Norman DC, Yoshikawa TT. Fever in the elderly. Infect Dis Clin North Am 1996;10(1):93–9. 5. Tal S, Guller V, Gurevich A, Levi S. Fever of unknown origin in the elderly. J Intern Med 2002;252(4):295–304. 6. Knockaert DC, Vanneste LJ, Bobbaers HJ. Fever of unknown origin in elderly patients. J Am Geriatr Soc 1993;41(11): 1187–92. 7. Arnow PM, Flaherty JP. Fever of unknown origin. Lancet 1997;350(9077):575–80. 8. Tabak F, Mert A, Celik AD, Ozaras R, Altiparmak MR, Ozturk R, et al. Fever of unknown origin in Turkey. Infection 2003;31(6):417–20. 9. Saltoglu N, Tasova Y, Midikli D, Aksu HS, Sanli A, Dundar IH. Fever of unknown origin in Turkey: Evaluation of 87 cases during a nine-year-period of study. J Infect 2004;48(1):81–5. 10. Kejariwal D, Sarkar N, Chakraborti SK, Agarwal V, Roy S. Pyrexia of unknown origin: A prospective study of 100 cases. J Postgrad Med 2001;47(2):104–7. 11. Ma XJ, Wang AX, Deng GH, Sheng RY. A clinical review of 449 cases with fever of unknown origin. Zhonghua Nei Ke Za Zhi 2004;43(9):682–5. 12. Barrier J, Schneebeli S, Peltier P. Les fievres prolongees inexpliquees chez les personnes, agees. Concours Med 1982; 104:4679–89. 13. Ramos Rincon JM, Ramos Guevara R, Herrero Huerta F. Fever of unknown origin in internal medicine. The experience of
16.
17.
18.
19.
20.
21. 22.
23.
24.
25. 26.
Spanish authors over 20 years. An Med Interna 1997;14(11): 585–92. ´smail Y. Tuberculosis—are we missing the diagnosis? Y Singapore Med J 2002;43(4):172–6. Boren E, Gershwin ME. Inflamm-aging: Autoimmunity, and the immune-risk phenotype. Autoimmun Rev 2004;3(5): 401–6. Zhiyong Z, Bingjun L, Xiaoju L, Xinjian F, Ping F, Wenya W. Fever of unknown origin: A report from China of 208 cases. Int J Clin Pract 2003;57(7):592–6. de Kleijn EM, van Lier HJ, van der Meer JW. Fever of unknown origin (FUO): II. Diagnostic procedures in a prospective multicenter study of 167 patients.The Netherlands FUO Study Group. Medicine (Baltimore) 1997;76(6): 401–14. de Kleijn EM, van der Meer JW. Fever of unknown origin (FUO): Report on 53 patients in a Dutch university hospital. Neth J Med 1995;47(2):54–60. Deng GH, Wang AX. Clinical analysis of 130 patients with fever of unknown origin. Zhonghua Nei Ke Za Zhi 1991;30(3): 157–9 [also see p. 188–189]. Knockaert DC, Vanneste LJ, Vanneste SB, Bobbaers HJ. Fever of unknown origin in the 1980s. An update of the diagnostic spectrum. Arch Intern Med 1992;152(1):51–5. Norman DC. Fever in the elderly. Clin Infect Dis 2000;31: 148–51. Knockaert DC, Dujardin KS, Bobbaers HJ. Long term followup of patients with undiagnosed fever of unknown origin. Arch Intern Med 1996;156(6):618–20. Quinn MJ, Sheedy 2nd PF, Stephens DH, Hattery RR. Computed tomography of the abdomen in evaluation of patients with fever of unknown origin. Radiology 1980; 136(2):407–11. Mourad O, Palda V, Detsky AS. A comprehensive evidencebased approach to fever of unknown origin. Arch Intern Med 2003;163(5):545–51. Arnow PM, Flaherty JP. Fever of unknown origin. Lancet 1997;350(9077):575–80. Ozaras R, Celik AD, Zengin K, Mert A, Ozturk KR, Cicek Y, et al. Is laparotomy necesssary in the diagnosis of fever of unknown origin? Acta Chir Belg 2005;105(1):89–92.