TUBERCULOSIS: THE MOST COMMON CAUSE OF FEVER OF UNDETERMINED ORIGIN IN SHIRAZ

A. Farhadi MD*, A. Parhizagar MD**, S. Rahman MD***, M. Haghshenas MD***

*Department of Internal Medicine, Mazandaran University of Medical Sciences. **Department of Internal Medicine, Iran University of Medical Sciences, *** Division of Hematology &Oncology Department of Internal Medicine, Shiraz University of Medical Sciences.

Background- Tuberculosis continues to be a common disease in our region and one of its unusual presentation is fever of undetermined origin (FUO).
Methods- During a 5-year prospective study on FUO in Shiraz University Hospitals, we investigated 68 cases of FUO, based on the criteria of Petersdorf and Beeson.
Results- Infections comprised the majority of the cases (44%), followed by neoplasm (19%), collagen vascular disease (15%), miscellaneous (9%) and undiagnosed (13%). Tuberculosis (TB) turned out to be the most common cause of fever of undetermined origin in this study (16.2%) and it comprised 37% of patients in the infectious group. The majority of the patients had extrapulmonary tuberculosis (9 out of 11).
Conclusion- As we expected tuberculosis is an important cause of FUO in our region where tuberculosis is still endemic. This is also true in other countries where the incidence of this disease has declined significantly in recent decades.

Keywords: FUO · tuberculosis

Introduction

Tuberculosis (TB) is still a common disease in our region, with an annual incidence of 30/100000.1 Case rates vary markedly according to age, sex, race and geographic location. Socioeconomic factors are also important determinants in the prevalence of TB.2

Annually 400 new cases of TB are registered in the tuberculosis control center of Fars province, central Iran, one fourth of which are extra pulmonary in origin.3 Tuberculosis was encountered as a common cause of fever of undetermined origin in several FUO series.4-13

While FUO is not a usual presentation of TB, it occurs occasionally as an intriguing problem in diagnosis. This is particularly true in the case of extrapulmonary TB. In this study we intended to determine different causes of FUO in a prospective

study. Emphasis is given to TB as a common cause of FUO in the region. Detecting this treatable cause of FUO more promptly and precisely would assist physicians in better management of FUO patients.

Materials and Methods

At the end of a 5-year period prospective study in Shiraz university of Medical Sciences, we documented 68 cases of fever of undetermined origin. The patients were chosen according to the criteria proposed by Petersdorf and Beeson (duration of disease exceeding 3 weeks, documented fever greater than 38.8 ?C on several occasions and no definite diagnosis after 1 week of in-patient evaluation).4,9 All patients were monitored closely, by repeated physical examinations. Several paraclinical work-ups were conducted during their admission and for at least 6 months after being discharged as out - patients.

Those patients that were labeled as FUO were evaluated until the diagnosis became evident. In the majority of the cases, there was no evidence of an active TB infection on admission or when the diagnosis was made. Most of our patients had received BCG vaccination in the past, therefore, no TB skin test was performed.11 Furthermore, there is a high rate of false positive TB skin test in the healthy population and high rate of asymptomatic infection with Mycobacterium tuberculosis or atypical mycobactera in our region. In fact, this renders interpretation of a positive test difficult, especially in a case with fever where other symptoms of an active TB infection are not present.

The diagnosis of patients was made by histopathology, staining body fluids for acid fast bacilli (AFB) or by radiographical findings and confirmation of the diagnosis was also made by the means of the therapeutic response to anti-TB medications.

Unfortunately, culture for Mycobacterium tuberculosis was not routinely available for all of the patients, therefore it was not performed in most of our cases. Tissue obtained by needle biopsy of peritoneum and pleura were the main materials for pathological diagnosis. Laparascopic peritoneal biopsy or pleuroscopic pleural biopsy were not available during this study. In addition, liver and lymph node histology were diagnostic in 1 and 2 cases, respectively. All sections were also stained for AFB in those showing granuloma.

Results

In our series of 68 cases of patients were divided into several groups according to their diagnosis. The infectious group comprised 44% of the cases followed by neoplastic disease (19%), collagen vascular disorders (15%) and miscellaneous disorders (9%). Thirteen percent were not diagnosed. In this study, 11 cases had TB (16.2%), which comprised 37% of FUO due to infection. The other infectious etiologies of FUO are depicted in Table 1. Six cases were male. The age range of patients was between 14 to 62, the mean being 40 years old. Nine out of 11 had extrapulmonary TB. The mean duration of illnesses prior to admission was 40 days, and the mean duration of hospitalization prior to diagnosis was 27 days. The illness started abruptly in one third of the cases. The pattern of fever was not diagnostic and it was remittent, intermittent and hectic in 6, 4 and 1 cases, respectively. True shaking chills were observed in 6 cases.

Peritoneum was the most common site of involvement (5 cases), followed by pleura (4 cases). The remaining cases were pulmonary (2 cases) lymphnode (2 cases), miliary (one case) and genitourinary (one case). Concomitant involvement of pleura and peritoneum were observed in 2 cases and peritoneum and lymph nodes in 2 other cases. From 4 pleural needle biopsies, 2 showed non specific inflammation, the third disclosed caseating granuloma and the fourth showed non-caseating granuloma which later proved to be due to TB in response to anti-TB medication. In all of these cases, thoracentesis was negative for AFB after several attempts. In addition AFB was not detected in special stains conducted to detect this organism in tissue sections.

Ascites was detected in those 5 patients with peritoneal involvement, however, only 3 out of 5 peritoneal needle biopsies showed caseating granuloma .

Pulmonary TB was diagnosed in 2 cases. The first was a young lady with minimal lung infiltration, whose diagnosis was made by using bronchoalveolar lavage for AFB staining. The second was an old lady who had apical lesions.The diagnosis was established only after treating empirically for reactive tuberculosis. TB lymph-adenitis was demonstrated pathologically in two cases in which involvement of the lymph nodes was associated with involvement of the peritoneum. The first case was a 14-year-old girl who developed chylous ascites and cervical lymphadenopathy whilst an in-patient. In addition, abdominal sonography and bipedal lymphangiography revealed retroperitoneal lymphadenopathy. The peritoneal biopsy was inconclusive and only revealed nonspecific inflammation. However cervical lymph node biopsy revealed caseating granuloma.

The second case was an old lady who was a known case of regularly dialyzed chronic renal failure, who presented with weight loss, and malaise. Ascetic fluid was exudative and the peritoneal biopsy revealed nonspecific inflammation. Abdominal CT scan showed a 2 by 3 centimeter, retroperitoneal lymph node, which contained caseating granuloma on sectioning after diagnostic laparotomy. In this patient, ascites disappeared after anti-TB therapy. AFB was not detected in ascetic fluid or lymph nodes in either cases.

Genitourinary TB was diagnosed in a patient with chronic renal failure, with detection of AFB in urine samples collected over a period of 24 hours. A 40-year-old man with miliary tuberculosis was diagnosed when noncaseating granuloma in bone marrow and caseating granuloma in the liver were detected by needle biopsy. It is noteworthy that the miliary pattern of chest X-ray was detected after meticulous re-evaluation of his film only after the diagnosis was made by the other methods.Marrow aspiration and biopsy performed in 9 patients were unremarkable except for one which showed noncaseating granuloma in the case of miliary tuberculosis. Abnormal alkaline phosphatase, associated with normal transaminase and bilirubine levels, was detected in the majority of our patients (7 cases).

While liver histologies were normal in 5 and only nonspecifically abnormal in 2 cases, it is interesting to note that the level of alkaline phosphatase was normal in the case of miliary tuberculosis. Thoracic and abdominal CT scans were performed in 2 cases, and both confirmed correct diagnosis.

Discussion

It is not surprising that tuberculosis is the most common cause of fever of undetermined origin, in this region, where the disease is still endemic. Our results are similar to those of other FUO investigators (Table 2).4,11

In our study, peritoneal tuberculosis was the most common type of TB producing FUO.

Although extrapulmonary tuberculosis is the most common manifestation of this infection as an FUO14, peritoneum is not a common site of involvement in other FUO series. Presentation of peritoneal tuberculosis as prolonged fever was also reported by other FUO investigators.15,16 Associated pleuropulmonary involvement, which was seen in 2 of our patients has been present in 25-83 percent of cases in literature.17-22 This suggests the possibility of hematogenous spread of the organism to or from the peritoneum.20

Levin recommend that the abdomen should be examined carefully in all patients who appear to have tuberculous pleuritis. If ascetic fluid is found, peritoneal biopsy appears to be preferable to pleural biopsy.18 Our experience supports this theory. Two afore mentioned patients had undergone pleural biopsy, which revealed non-specific inflammatory reaction, whilst peritoneal biopsy at a later date, led to correct diagnosis. On the other hand as noted by Gonnella: the presence of an abnormal chest X-ray in a patient with ascites should alert one to the possibility of TB peritonitis.

This is particularly true of pleural effusion.17 Recently, using laparascopic biopsy of the peritoneum has replaced percutaneous peritoneal biopsy in cases with lymph dominant ascites and fever with near 100% sensitivity.23 The role of determination of ascetic fluid adenosine deaminase (ADA) in the diagnosis of TB peritonitis is unclear.24

Tuberculosis of the lung produces no difficulty in diagnosis in the majority of cases. However, active pulmonary tuberculosis can be present in normal chest X-rays.25 In one of our cases, minimal lung infiltration had gone unnoticed untill some time later, as an in-patient bronchoalveolar lavage became positive for AFB.

The other patient had apical calcification on her chest CT scan that was not noticed in the routine chest X-ray. This case was treated empirically for possible reactivated pulmonary tuberculosis, she responded well to the therapy with restoration of appetite and weight, increase in hemoglobin level and decrease in erythrocyte sedimentation rate.

Two patients had tuberculus retroperitoneal lymphadenitis associated with peritonitis. Lymph node biopsies showed caseating granuloma without AFB on sectioning, while peritoneal biopsies yielded nonspecific inflammation.

In two of our cases, chronic renal failure acted as predisposing factor of TB of the genitourinary tract and TB lymphadenitis associated with peritonitis. The increase in the susceptibility of patients with chronic renal failure to tuberculosis has also been noted in other reports.12

The role of miliary tuberculosis in producing F.U.O is known and has been observed in several reports.4,10 In the majority of these cases, the older patients were more affected with prolong illness.

Our case was a 40-year-old man who had fever, cough and malaise for 1 month prior to admission. The bone marrow examination and liver biopsy were diagnostic in this case, however these procedures were usually unremarkable or nonspecifically abnormal in the remainder of the cases.

Tuberculosis is one of the differential diagnoses of elevated alkaline phosphatase in F.U.O patients.26 However, this finding is very nonspecific, and as is evident from our series, it is usually associated with normal or nonspecifically abnormal histology.

The prevalence of TB had declined significantly in developing countries, but since 1985 this trend has reversed.27 The change in the case rate of TB has been due almost entirely to an increase in the number of cases of pulmonary diseases, and there has been little changes in the average number of extra-pulmonary cases reported.28

Hence, F.U.O is mainly the presentation of extra pulmonary TB; this is why the disease is still a common cause of F.U.O even in non-endemic areas. This is especially true for the fever of undetermined origin in patients infected with the human immunodeficiency virus.29 So it is concluded that hidden tuberculosis presented as FUO is a world wide problem in medical practice.

References

  1. Personal communication with Tuberculosis control center. Tehran, Iran, 1993.
  2. Glassroth J, Robins AG, Snider JR. Tuberculosis in 1980s. N Engl J Med 1980; 302: 1441-50.
  3. Personal communication with Fars Tuberculosis control center. Shiraz, Iran, 1993.
  4. Pedersdorf RG, Beeson PB. Fever of unexplained origin: Report of 100 cases. Med 1961; 40; 1.
  5. Bursch JL, Weinstein L. fever of unknown origin. Med Clin North Am. 1988 sep 72 (5): 1247-61.
  6. Jacoby GA, Swartz MN. Fever of undetermined origin. N Engl J Med 1973; 289: 1407-10.
  7. Molavi A, Weinstein L. Persistent perplexing pyrexia: Some comments on etiology and diagnosis. Med Clin North Am 1970; 5(6): 379-96.
  8. Petterson T. Fever of obscure origin: A follow-up investigation of 88 cases. Acta Med Scan 1962; 171: 575.
  9. Petersdorf RG. F. U. O: How it has changed in 20 years. Hosp Prac April 1985; 15; 20(4) 841-84.
  10. Larson EB, Featherstone HJ, Petersdorf RG. Fever of undetermined origin. Diagnosis and follow-up of 105 cases; 1970-1980. Med 1982; 61: 269-22.
  11. Barbads. FJ, Vazquez JJ, Pena JM, ET al. Fever of undetermined origin: A survey on 133 patients. J Med 1984; 15: 185-22.
  12. Smith JW. Fever of undetermined origin: Not what it used to be. Am J Med Sci 1986; 292:56-64.
  13. Dinarello CA, Wolff SM. Fever of unknown origin. In; Mandel G, Douglas RG, bennett JE, eds. Principals and Practice of Infectious Diseases. New York, Churchill livingstone, 1990.
  14. Verden SG, Berenninkmeijer BJ, Vandermeer JW. Peritoneal tuberculosis in two young immigrants with fever of undetermined origin. Neth J Med 1992; 41:218.
  15. Rakatansky H, Kirsner JB. The gastrointestinal tract: An often forgotten source of prolonged fever. Arch Intern Med 1967; 119: 321-8.
  16. Gonnella JS, Hudson EK. Clinical pattern of tuberculosis peritonitis. Arch Intern Med 1966; 117: 164 -9.
  17. Levin H. Needle biopsy in diagnosis of tuberculous peritonitis. Am Rev Resp Dis 1968; 97: 889.
  18. Borhanmanesh F, Hekmat K, Vaez-zadeh K, et al.. Tuberculous peritonitis: Prospective study of 32 case in Iran. Ann Int Med 1972; 76: 567-72.
  19. Burack WR, Hollister RM. Tuberculous peritonitis: A study of 47 proved cases encountered by a general medical unit in 25 years. Am J Med 1980; 28:510.
  20. Bastani B, Shariat-zadeh MR, Dehdashti F. Tuberculous peritonitis: Report of 30 cases and review of literature. Puot J Med, New Series 1985; 56: 549-57.
  21. Karney WW, O’Donoghue JM, Ostrow JH, et al.. The spectrum of tuberculous peritonitis. Chest 1977; 72: 310-5.
  22. Reddy KR, Diprima RE, Raskin, JB, et al. Tuberculous peritonitis: Laparascopic diagnosis of an uncommon disease in United States. Gastrointest Endosc 1988; 34: 442-6.
  23. Hillebrand Dj, Runyon BA, Yasmineh WG, et al. Ascetic fluids adenosine deaminase insensitivity in detecting
  24. tuberculous peritonitis. Hepatology 1996, 24, 1408-12.
  25. Husen L, Falkerson LL, Vecchio E, et al. Pulmonary tuberculosis with negative finding on chest X-ray films: A study of 40 cases. Chest 1971; 60: 540-2.
  26. Raviglione MC, O’Brien RJ. Tuberculosis. In: Fauci AS, Braunwald E, Isselbacher K, et al, eds. Harrison’s Principles of Internal Medicine. New York: McGraww-Hill, 1998
  27. Cunha BA. Fever of unknown origin in the elderly. Geriatrics 1982; 37:30-6, 39-40, 44.
  28. Brundney K, Dobkin J. Resurgent of tuberculosis in New York city: Human immunodeficiency virus, Homeless-ness, and the decline of tuberculosis control programs. Am Rev Respire Ids 1991; 144: 745.
  29. Fairer LS, Lowed AM, Meadow MAP. Extra pulmonary tuberculosis in United State. AM J Epidemiol 1979; 109: 205.
  30. Miralles P, Moreno S, prez Tascon M, et al. Fever of uncertain origin in patients infected with the human immunodeficiency virus. Clin Infect Dis 1995; 20: 827.

    AIM Home|Table of Contents