Familial Mediterranean fever

Family physicians may commonly encounter patients with an acute febrile illness of less than two weeks' duration. Often the illness is thought to be viral in origin, and generally the patient recovers quickly with no morbid sequelae. In some patients, however, the fever persists, causing discomfort and alarm to both the patient and the physician.

The term fever of unknown origin (FUO) is reserved for elevations of temperature greater than 38.3[degrees]C (101.0[degrees]F) occurring for prolonged periods (at least two to three weeks), for which a cause cannot be confidently determined despite at least one week of intensive study.(1)(2)(3)(4)

Some patients with documented FUO remain undiagnosed despite extensive evaluation and may do well clinically (for example, patients with a prolonged viral syndrome). Frequently, however, the fever is a harbinger of a more significant underlying pathology, such as malignancy or serious, deep-seated infection. For this reason, a thorough but directed evaluation is indicated in patients with FUO.(1)(2)(3)(4) Empiric antibiotic therapy before adequate diagnostic evaluation is seldom justified.(5)

A wide spectrum of disease entities may be responsible for FUO. These may be broadly classified as follows:

Infection. Infections are responsible for approximately 20 to 30 percent of cases of FUO.(2)(3)(4) Fungal infections, sinusitis, chronic osteomyelitis, chronic persistent viral infections such as human immunodeficiency syndrome (HIV), and soft tissue abscesses (seen more commonly in immunocompromised patients or in patients with a history of previous surgical instrumentation) are common examples.

Neoplasms. Neoplasms are responsible for 15 to 25 percent of cases of FUO.(2)(3)(4) Hematologic tumors such as leukemias and Hodgkin's and non-Hodgkin's lymphomas may present with persistent fever. Fever is less likely to be a presenting symptom in patients with solid tumors. However, fever may occur in patients with solid tumors that obstruct ductal systems (leading to secondary infection) or in patients with tumors that infiltrate bone marrow.

Collagen Vascular Diseases. Collagen vascular diseases account for 15 percent of cases of FUO and are more prevalent among women.(2)(3)(4) These disorders, such as systemic lupus erythematosis, rheumatoid arthritis, polyarteritis nodosa and Wegener's granulomatosis, often involve multiple organ systems.

Granulomatous Entities. Granulomatous disease states such as sarcoidosis, Crohn's disease, erythema nodosum and chronic infection with granulomas resulting from fungi and tuberculosis are another cause of FUO. Like collagen vascular diseases, many granulomatous disease states affect multiple organs. Active granulomatous entities, rather than old, healed granulomas, are important causes of FUO.

Miscellaneous Other Causes. Miscellaneous causes include drug fever, multiple pulmonary emboli, cerebrovascular accident, factitious fever (created by patients through the self-administration of pyrogenic substances such as feces, milk or urine) and metabolic diseases like familial Mediterranean fever and hyperthyroidism. Granulomatous and miscellaneous causes contribute the remaining 15 percent of cases of FUO.(2)(3)(4)

Evaluation

Given the multitude of potential etiologies for fever, each evaluation should be individually tailored. A careful history (including questions about foreign travel, potential infectious exposure, family history, diet and medications) and consideration of patient age, as well as a detailed physical examination and screening laboratory studies (i.e., complete blood count, serum chemistries, liver function tests, antinuclear antibody, rheumatoid factor and erythrocyte sedimentation rate) are important factors that may steer the clinician toward a particular category of FUO.

If the history or physical signs enable a potential site to be localized (e.g., local pain, a palpable mass, a history of recurrent focal infection or recent invasive procedure), anatomic imaging of the suspected region should be undertaken. This imaging may include plain radiographs, ultrasound examination, computed tomographic (CT) scans or magnetic resonance imaging (MRI).(6) If these studies fail to detect a cause for the FUO, or if the patient's history and physical examination fail to localize an abnormality, it may be more cost-effective to perform whole-body imaging using scintigraphy than to continue imaging individual areas.(6)(7)(8)(9)(10)

Nuclear Imaging Techniques

Several nuclear medicine methods may aid in the localization of a cause for FUO. These techniques have been shown to be most effective in the evaluation of occult infections, but they also have a limited role in tumor imaging.(6)(7)(8)(9)(10) Rather than directly detecting the presence of infection, they depict the inflammatory response.

BONE SCANNING

Three-phase bone scanning using technetium Tc 99m is most useful in the evaluation of suspected occult osteomyelitis (Figure 1).

[CHART OMITTED]

Reported sensitivity of bone scintigraphy for osteomyelitis varies from 32 percent to 100 percent, with lower percentages in neonates and patients with severe osteoporosis or severe peripheral vascular disease. For acute osteomyelitis, sensitivity is typically in the 90 percent range.(11)(12)

In general, it is the specificity of the bone scan, not the sensitivity, that is the issue. Increased tracer uptake in bone may result from noninfectious causes of increased metabolic activity, such as healing fracture, heterotopic bone formation or degenerative joint disease.

Since reported specificity may range from zero percent to 100 percent, positive bone scans must be interpreted with regard to the individual clinical setting. For example, low specificity is often encountered in the feet of diabetic patients and at sites of previous surgery. In addition, the finding of nonspecific abnormal tracer accumulation on the bone scan corresponding to a nonosseous anatomic location may lead to the diagnosis of soft tissue infection(11)(12)(13) (Figure 2). Because the tracer localizes mainly in osseous structures, bone scanning is insensitive in diagnosing soft tissue causes for FUO. Consequently, if soft tissue pathology is suspected, bone scanning is less useful than gallium- or indium-labeled leukocyte scanning.(9) The bone scan may occasionally detect undiagnosed osseous metastatic tumor in a patient with fever, with or without bone pain.

[CHART OMITTED]

GALLIUM SCANNING

Gallium imaging is more useful than bone scanning in the evaluation of FUO.(14) Gallium 67 behaves somewhat like a ferric ion analog and, after intravenous injection, is rapidly bound to serum proteins such as transferin and lactoferrin. Uptake at sites of infection has been well documented, and several mechanisms for this process have been postulated. Bacteria have high iron requirements and thus take up gallium avidly by means of specialized iron-binding proteins. Leukocyte lactoferrins--proteins present extracellularly at sites of infection--are also thought to bind well to gallium.(14)

Gallium uptake has been well demon-strated in a wide variety of soft tissue infections, deep-seated abscesses, osteomyelitis, pulmonary infections (including Pneumocystis carinii pneumonia)(15)(16) and, in addition, in many solid tumors, including Hodgkin's and non-Hodgkin's lymphoma, lung and breast carcinoma, malignant melanoma and malignant thymoma(14)(17)(18) (Figures 3 and 4). The mechanism for tumor uptake remains obscure, but many tumors express ferritin, to which(67) Ga may bind.(14) Because gallium binds well to both inflammation and tumor, the usefulness of the scan lies in localization of the problem area, rather than in definitive diagnosis of the abnormality. The diagnosis is usually confirmed by directed biopsy, cytology and culture.

[CHART OMITTED]

INDIUM-LABELED LEUKOCYTE SCANNING

Indium 111-labeled autologous leukocytes were developed on the premise that, in the presence of an inflammatory process, freely circulating, viable [111.sup.In-labeled] autologous leukocytes will be incorporated into the inflammatory process in the same manner as nonlabeled leukocytes.(7)(8) The technique requires withdrawing approximately 50 mL of autologous whole blood, separating the leukocytes from the blood, labeling them with [111.sup.In] and reinjecting the labeled leukocytes.

These scans are often used in conjunction with [99m.sup.Tc] bone scans to improve the specificity of diagnosing osteomyelitis in the face of trauma or degenerative joint disease(11)(12)(13) (Figure 5). Indium-labeled white blood cell scans are also used to evaluate the current activity and extent of inflammatory bowel disease in patients with an established diagnosis. The scans are also useful in localizing areas of occult infection or abscess throughout the body(7)(8) (Figure 6).

[CHART OMITTED]

In general, [111.sup.In-labeled] leukocyte scanning is more useful when the duration of the suspected infectious process is three weeks or less; gallium is used in more long-term infections.(7) However, some controversy surrounds this premise.(19)(20)

In general, [111.sup.In-labeled] leukocyte scanning is the preferred method for imaging the abdomen because, unlike gallium, [111.sup.In-labeled] leukocytes are not normally excreted in the bowel. Gallium may be more useful in imaging for chest infections and for evaluating the possibility of vertebral osteomyelitis and diskitis.(13)(14) Gallium has a further advantage in cases of FUO because of its ability to localize both infection and tumor. However, the normal excretion of gallium in the bowel makes interpretation of abdominal scans difficult, especially after abdominal surgery.(6)(14)

If abdominal pathology is suspected, delayed imaging over several days and the use of laxatives to expel colonic contents may be necessary before the persistence of abnormal nonalimentary abdominal gallium uptake can be confirmed.(14) The sensitivity, specificity and other characteristics of several scintigraphic techniques are outlined in Table 1.

[TABULAR DATA OMITTED]

Illustrative Case

A 74-year old man presented at a hospital with a two-month history of spiking fevers and chills, a weight loss of 30 kg (65 1b) and generalized weakness. His past surgical history included an abdominal aortic aneurysm repair by graft. Except for cachexia and pyrexia, findings on physical examination were unremarkable. On two separate occasions, blood cultures were positive for methicillin-sensitive Staphylococcus aureus and Escherichia coli.

The patient was treated with intravenous antibiotics for six weeks without defervescence. Because the infection had no obvious site of origin, an [111.sup.In-labeled] leukocyte scan was performed. Planar images demonstrated a focus of abnormal increased activity anterior to the lumbosacral spine, in the region of the abdominal aortic aneurysm graft, a pattern consistent with an infected aortic graft (Figure 7). Contrast CT scan of the abdomen and laparotomy confirmed an infected graft. The graft was replaced, and the patient was treated with several more weeks of intravenous antibiotics. His fever defervesced completely and he continued to do well.

[CHART OMITTED]

Use in Opportunistic Infections

Opportunistic infections in patients with HIV infection deserve special attention, because nuclear medicine techniques may be particularly useful in these patients. A negative chest radiograph does not exclude the diagnosis of P. carinii pneumonia, since 5 to 10 percent of patients subsequently prove to have the infection despite a normal chest radiograph.(21)

Gallium binds well in the presence of P. carinii pneumonia, with a sensitivity and a specificity of 80 to 90 percent and 50 to 74 percent, respectively; specificity increases to nearly 100 percent in patients with a normal chest radiograph.(15) Gallium does not localize in Kaposi's sarcoma. Therefore, in a patient with acquired immunodeficiency syndrome who has an abnormal chest radiograph and a negative gallium scan, Kaposi's sarcoma is the likely diagnosis.

Another useful and innovative nuclear medicine procedure in patients with P. carinii pneumonia is the [sup.99m.Tc] DTPA (diethylenetriamine pentaacetic acid) transfer study of the lungs, which involves the inhalation of a radioaerosol and the measurement of rate of clearance from the lungs. An abnormal result (i.e., a biphasic curve with a rapid first component rather than a single exponential curve) is consistent with alveolitis. The clearance rate of the aerosol is higher for P. carinii pneumonia than for other pulmonary conditions.

Patients with AIDS have an increased incidence of gastric and intestinal lymphoma and enteric infections and, although evaluation of the abdomen with gallium is somewhat limited as described above, it may have a role in the detection of abdominal (and extra-abdominal) lymphoma and occult enteritis, colitis and perirectal abscesses.(21)

Because none of the established nuclear medicine techniques are ideal, a number of new and currently experimental techniques are presently under investigation, although none have yet been approved by the U.S. Food and Drug Administration for routine clinical use. These approaches include [sup.99m.Tc] HM-PAO (hexamethyl-propyleneamine-oxime) labeling of white blood cells in vitro, which still requires handling of blood outside of the body, but because the imaging characteristics of technetium are greatly superior to those of indium, improved images may be obtained.(22)

Radiolabeled monoclonal antibodies may be directed to granulocytes and other white blood cells in vivo.(10)(23) One advantage of this technique is that white blood cell imaging is achieved by in vivo labeling without necessitating the withdrawal of blood. Nonspecific polyclonal human gamma globulin labeled with [sup.111.In] has also been shown to concentrate at sites of inflammation(24)(25)(26) (e.g., in the lungs in cases of P. carinii pneumonia).(25) Microcolloids or serum albumin labeled with [sup.99m.Tc] may also diffuse into sites of inflammation, possibly because of the increased permeability of inflamed capillaries.(9)(25) Finally, radiolabeled lymphokines, hematopoietic hormones and inflammatory mediators may provide noninvasive, in vivo probes of infectious and inflammatory processes.(27)(28)

Clinical Algorithm for FUO Patients

Table 2 describes the steps for evaluation of patients with FUO. Because the exact choice of modality depends on the particular equipment and skills that are locally available, this approach should be viewed as a suggested protocol, rather than a clinical standard of care.

TABLE 2 Steps in the Investigation of Patients with Fever of Unknown

Origin

Exclude factitious fever.

Take a careful history and do a detailed physical examination, followed by screening laboratory studies that include complete blood cell count, serum chemistries, liver function tests, antinuclear antibody, rheumatoid factor and erythrocyte sedimentation rate.

Obtain serial aerobic and anaerobic blood cultures (no more than six sets of cultures); culture urine if leukocytes are present; consider fungal blood cultures.

Obtain baseline chest radiograph.

If specific anatomic location is suspected (based on history or physical examination), obtain anatomic imaging study of the area in question; if osteomyelitis is suspected and plain radiograph is negative, obtain three-phase technetium 99m bone scan.

If diagnosis is still negative or if the fever is accompanied by no localizing features, consider whole-body nuclear medicine imaging procedure: gallium-67 scans may be superior to indium-111-labeled autologous leukocyte scans for chronic infections and undiagnosed malignancies, but its usefulness is somewhat limited in the abdomen.

Figure 6 from Fig LM, Shulkin BL. Scintigraphic detection of sinusitis. Eur J Nucl Med 1988; 14:552-4. Used with permission.

REFERENCES

(1.)Larson EB, Featherstone HJ, Petersdorf RG. Fever of undetermined origin: diagnosis and follow-up of 105 cases, 1970-1980. Medicine 1982; 61:269-92.

(2.)Jacoby GA, Swartz MN. Fever of undetermined origin. N Engl J Med 1973; 289:1407-10.

(3.)Root RK, Petersdorf RG. Chills and fever. In: Wilson JD, ed. Harrison's Principles of internal medicine. 12th ed. New York: McGraw-Hill, 1991:125-33.

(4.)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; 152:51-5.

(5.)DiNubile MJ. Acute fevers of unknown origin. A plea for restraint. Arch Intern Med 1993; 153:2525-6.

(6.)McNeil BJ, Sanders R, Alderson PO, Hessel SJ, Finberg H, Siegelman SS, et al. A prospective study of computed tomography, ultrasound, and gallium imaging in patients with fever. Radiology 1981; 139:647-53.

(7.)Schauwecker DS, Burt RW, Park HM, Mock BH, Tobolski MM, Yu PL, et al. Comparison of purified indium-111 granulocytes and indium-111 mixed leukocytes for imaging of infections. J Nucl Med 1988; 29:23-5.

(8.)Sfakianakis GN, Al-Sheikh W, Heal A, Rodman G, Zeppa R, Serafini A. Comparisons of scintigraphy with In-111 leukocytes and Ga-67 in the diagnosis of occult sepsis. J Nucl Med 1982; 23:618-26.

(9.)Datz FL, Morton KA. Radionuclide detection of occult infection: current strategies. Cancer Invest 1991; 9:691-8.

(10.)Becker W, Dolkemeyer U, Gramatzki M, Schneider MU, Scheele J, Wolf F. Use of immunoscintigraphy in the diagnosis of fever of unknown origin. Eur J Nucl Med 1993; 20:1078-83.

(11.)Gupta NC, Prezio JA. Radionuclide imaging in osteomyelitis. Semin Nucl Med 1988; 18:287-99.

(12.)Schauwecker DS, Braunstein EM, Wheat LJ. Diagnostic imaging of osteomyelitis. Infect Dis Clin North Am 1990; 4:441-63.

(13.)Wegener WA, Alavi A. Diagnostic imaging of musculoskeletal infection. Roentgenography; gallium, indium-labeled white blood cell, gammaglobulin, bone scintigraphy; and MRI. Orthop Clin North Am 1991; 22:401-18.

(14.)Neumann RD, Hoffer PB. Gallium-67 scintigraphy for detection of inflammation and tumors. In: Freeman LM, ed. Freeman and Johnson's Clinical radionuclide imaging. 3d ed. Orlando: Grune & Stratton, 1984:1319-64.

(15.)Tumeh SS, Belville JS, Pugatch R, McNeil BJ. Ga-67 scintigraphy and computed tomography in the diagnosis of Pneumocystis carinii pneumonia in patients with AIDS. A prospective comparison. Clin Nucl Med 1992; 17:387-94.

(16.)Barron TF, Birnbaum NS, Shane LB, Goldsmith SJ, Rosen MJ. Pneumocystis carinii pneumonia studied by gallium-67 scanning. Radiology 1985; 154:791-3.

(17.)Bekerman C, Hoffer PB, Bitran JD. The role of gallium-67 in the clinical evaluation of cancer. Sem Nucl Med 1984; 14:296-323.

(18.)Chandramouly BS, Scagnelli T, Burgess CK. Uptake of gallium in the mediastinum. Semin Nucl Med 1989; 19:247-9.

(19.)Datz FL, Thorne DA. Effect of antibiotic therapy on the sensitivity of indium-111-labeled leukocyte scans. J Nucl Med 1986; 27:1849-53.

(20.)Chung CJ, Hicklin OA, Payan JM, Gordon L. Indium-111-labeled leukocyte scan in detection of synthetic vascular graft infection: the effect of antibiotic treatment. J Nucl Med 1991; 32:13-5.

(21.)O'Doherty MJ, Nunan TO. Nuclear medicine and AIDS. Nucl Med Commun 1993; 14:830-48.

(22.)Kao CH, Wang SJ. Spread of infectious complications of odontogenic abscess detected by technetium-99m-HMPAO-labeled WBC scan of occult sepsis in the intensive care unit. J Nucl Med 1992; 33:254-5.

(23.)Sciuk J, Brandau W, Vollet B, Stucker R, Erlemann R, Bartenstein P, et al. Comparison of technetium 99m polyclonal human immunoglobulin and technetium 99m monoclonal antibodies for imaging chronic osteomyelitis. First clinical results. Eur J Nucl Med 1991; 18:401-7.

(24.)Oyen WJ, Claessens RA, van der Meer JW, Rubin RH, Strauss HW, Corstens FH. Indium-111-labeled human nonspecific immunoglobulin G: a new radiopharmaceutical for imaging infectious and inflammatory foci. Clin Infect Dis 1992; 14:1110-8.

(25.)Rubin RH, Fischman AJ, Needleman M, Wilkinson R, Callahan RJ, Khaw BA, et al. Radiolabeled, non-specific, polyclonal human immunoglobulin in the detection of focal inflammation by scintigraphy: comparison with gallium-67 citrate and technetium-99m-labeled albumin. J Nucl Med 1989; 30:385-9.

(26.)Barrow SA, Graham W, Jyawook S, Dragotakes SC, Solomon HF, Babich JW, et al. Localization of indium-111-immunoglobulin G, technetium-99m-immunoglobulin G and indium-111-labeled white blood cells at sites of acute bacterial infection in rabbits. J Nucl Med 1993; 34:1975-9.

(27.)Fischman AJ, Pike MC, Kroon D, Fucello AJ, Rexinger D, ten Kate C, et al. Imaging focal sites of bacterial infection in rats with indium-111-labeled chemotactic peptide analogs. J Nucl Med 1991; 32:483-91.

(28.)Hay RV, Skinner RS, Newman OC, Kunkel SL, Lyle LR, Shapiro B, et al. Nuclear imaging of acute inflammatory lesions with recombinant human interleukin-8 [Abstract]. J Nucl Med 1993; 34:104.

COPYRIGHT 1994 American Academy of Family Physicians
COPYRIGHT 2004 Gale Group