Role of 18F-FDG PET/CT in Patients With Pyrexia Of Unknown Origin Academic research paper on "Clinical medicine"

Accepted Manuscript

Role of 18F-FDG PET/CT in Patients With Pyrexia Of Unknown Origin

Dr. Anuradha Rao, MBBS, DMRD, DNB, Dr. Kapil Shirodkar, MBBS, DMRD, EDIR, Dr. M.J. Govindarajan, MBBS, MD, Dr. Siddalinga Devaru, MBBS, DMRD, MD, DRM

PII: S2352-6211(16)30079-1

DOI: 10.1016/j.jrid.2016.11.006

Reference: JRID 104

To appear in: Radiology of Infectious Diseases

Received Date: 5 August 2016 Revised Date: 20 November 2016 Accepted Date: 20 November 2016

Please cite this article as: Rao A, Shirodkar K, Govindarajan MJ, Devaru S, Role of 18F-FDG PET/CT in Patients With Pyrexia Of Unknown Origin, Radiology of Infectious Diseases (2017), doi: 10.1016/ j.jrid.2016.11.006.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Role of 18F-FDG PET/CT in Patients With Pyrexia Of Unknown Origin

Keywords: PUO, Neoplasia, Inflammation, Infection, Biopsy, Ultrasound, PET-CT, CT, Nuclear medicine, Molecular imaging, Lymph nodes

Abbreviations: PUO: Pyrexia of unknown origin; TP: True positive, FP: False Positive; TN: True Negative; FN: False Negative

Abstract:

Objectives: This paper is a prospective observational study to assess the role of 18F-FDG PET/CT in evaluation of pyrexia of unknown origin (PUO) in a hospital based population in a tertiary level referral center in an Indian metropolitan city.

Materials and Methods: Medical records of 25 patients who underwent 18FDG PET/CT as a part of investigative work up of PUO were analyzed. A diagnostic CT scan was performed from the level of clavicle to the pubic symphysis in arterial and venous phases of contrast administration. Equilibrium phase images were obtained from vertex to mid thigh without breath hold instructions with subsequent PET acquisition from vertex to mid-thigh, and these two sets of images were fused to obtain PET/CT fused images. Two radiologists with considerable PET-CT experience evaluated the contrast enhanced CT and 18-FDG PET-CT images separately and tried to identify the probable cause of fever in the 25 patients.

Results: Records of 25 patients (13 Females, 12 Males) were evaluated. Abnormal focal FDG uptake was found in 21 patients and contributed to final diagnosis (TP) in 20 patients with 1 false positive results (FP). In the rest of the 4 patients with negative PET/CT final specific diagnosis (FN) was made only in 2 patients (2/4-50%) and in 2 patients no diagnosis was made (TN).The PET/CT had a sensitivity of 90.91%, specificity of 66.67%, positive predictive value(PPV) of95.24%and negative predictive value(NPV-50 %). TP cases included 8 cases of Kochs ( 2 cases of tubercular lymphadenitis, 1 case of tubercular pleural effusion, 1 case of spinal kochs, 1 case of genitourinary kochs, 1 case of Cutaneous Koch, 1 case of splenic Koch; 1 case of abdominal Koch); 4 cases of focal infections/abscesses, 1 cases of lymphoma, 2 cases of Sarcoidosis, 1 case of Kikuchi Fusimoto, 1 case of aortoarteritis, 1 case of Castleman's disease, 1 case of Rosai Dorfman and 1 case of thyroiditis. FN negative cases included one case of meningitis and one case of giant cell arteritis. One false positive case was due to focal hyper metabolism on PET in the caecum.

In comparison, on CECT, probable etiology was suspected in 11 patients, out of which 8 cases were true positive (TP), 3 cases were false positive and 12 were false negative. In 2 patients no diagnosis was made (TN). CECT had a sensitivity of 42.86%, specificity of 50% positive predictive value(PPV) of 84 %and negative predictive value (NPV-14.29 %). Thus 18F FDG PET/CT improved the sensitivity (90.91% vs 42.86 %),

specificity (66.67 % vs 50%), Positive predicate value (95.24% vs 84 %) and Negative predicate value (50 % vs 14.29 %) as compared CECT alone.

Conclusions:

FDG PET/CT is better than CECT in making a specific diagnosis in the evaluation of PUO. It also helps in deciding the best possible site for biopsy.

Authors:

1.Dr. Anuradha Rao1, MBBS, DMRD, DNB; 1 Department of Radiology, Apollo Hospitals, Bangalore, India, 560076

2.Dr. Kapil Shirodkar1, MBBS, DMRD, EDIR; 1 Department of Radiology, Apollo Hospitals, Bangalore, India, 560076

3.Dr. Govindarajan M J1, MBBS, MD; 1 Department of Radiology, Apollo Hospitals, Bangalore, India, 560076

4.Dr. Siddalinga Devaru1, MBBS, DMRD, MD, DRM, 1 Department of Radiology, Apollo Hospitals, Bangalore, India, 560076

Corresponding Author: Dr. Anuradha Rao, MBBS, DMRD, DNB1 Department of Radiology, Apollo Hospitals, Bangalore, India 560076; anu78rao@gmail.com

1. INTRODUCTION

Patients presenting to physicians with history of fever are usually readily diagnosed on the basis of presenting signs and symptoms. Though fever is a common clinical condition, it can sometimes be challenging when the patient does not respond appropriately despite standard antibiotics and antipyretics. Under these circumstances, an elaborate search for the underlying pathology is initiated.

Pyrexia of Unknown origin (PUO) or Fever of unknown origin (FUO) was first defined in 1961 by Petersdorf and Beeson as recurrent fever of 38.3 degree Celsius or higher, lasting 3 week or longer, and undiagnosed after1 week of hospital evaluation [1]. A new classification system was proposed by Durack and Street in 1991 [2] and divided PUO as Classic PUO, neutropenic PUO, nosocomial PUO and PUO associated with HIV infection. Roth et al [3] defined PUO as a temperature higher than 38.3°C on several occasions and the fever lasting more than three weeks, and a failure to reach a diagnosis despite one week of inpatient investigation. This strict definition restricts common and self-limiting medical conditions from being included as PUO. American Association of Family Physicians sub-classified PUO into 4 categories—classic, nosocomial, immune deficient, and HIV-associated. Each group has a unique differential diagnosis based on characteristics and vulnerabilities and therefore, a different process of evaluation. Hersch et al [4] classified PUO under the following subgroups-infection (20% to 40%), malignancy (20% to 30%), noninfectious inflammatory diseases (10% to 30%), miscellaneous (10% to 20%), and undiagnosed (up to 50%). Noninfectious inflammatory diseases commonly include connective tissue diseases, vasculitis, and granulomatous diseases. In developed countries, the noninfectious inflammatory diseases and undiagnosed groups comprise a higher proportion of PUO cases. Higher rates of infection and neoplasm are seen in underdeveloped countries. Drug fever is implicated in 1% to 3% of PUO cases. True PUO can be termed for those of about 9% - 30% of patients who go undiagnosed [5].

Endoscopy, biopsy, high-resolution CT, MRI, and nuclear medicine techniques are being used as the initial line of investigation. It has been long recognized that 18F-FDG accumulates not only in malignant tissues but also at sites of infection and inflammation and in autoimmune diseases. 18F-FDG is promising in patients with PUO as the causes most often responsible for the condition are usually FDG avid [5]. Hypermetabolic foci in areas of infection/ inflammation are identified much before the appearance of abnormal anatomical features by PET thereby helping in early diagnosis and early initiation of treatment [6] as well as in guiding biopsy. Various studies in the past have tried to define the role of FDG PET and PET/CT [7-11]. This paper is a prospective observational study to assess the role of 18F-FDG PET/CT in evaluation of PUO in an Indian metropolitan city.

2. MATERIALS AND METHODS

Medical records of 25 patients who underwent 18FDG PET/CT as a part of investigative work up of PUO were analyzed. PUO was defined as a fever of more than 38.3degree C that lasted for more than 3 weeks and failure to reach diagnosis after more than 1 week of inpatient investigation.

2.1 Imaging parameters of CECT and PET-CT

F-18 FDG Injection of 10- 13 millicuries was administered intravenously and the patient made to wait in a quiet waiting room without significant physical activities for 60-75 minutes after which scan was performed. A diagnostic CT scan was performed from the level of clavicle to the pubic symphysis in arterial and venous phases of contrast administration. Equilibrium phase images will be obtained from vertex to mid thigh without breath hold instructions with subsequent PET acquisition with about 60 seconds per bed position from vertex to mid-thigh and these two set of images were fused to obtain PET/CT fused images. CT morphology, enhancement patterns and FDG uptake in the form of Max SUV was considered for assessment of the images.

2.2 Image Interpretation

The images (fused18F-FDG PET/CT and CECT) were displayed on a standard PET/CT viewing workstation (Fusion Viewer, Philips Medical Systems, Netherland) with multiplanar capabilities. Two radiologists with considerable PET-CT experience evaluated the Contrast enhanced CT and 18-FDG PET/CT images separately and tried to identify probable cause of infection in the 25 patients. The radiologists identified probable cause of the PUO and also tried to identify the probable and most accessible site for biopsy.

3. RESULTS

Records of 25 patients (13 Females, 12 Males) of age group ranging from 16 - 72 years were evaluated. Histopathogy was considered as the gold standard for final diagnosis wherever appropriate. Abnormal focal FDG uptake was found in 21 patients and contributed to final diagnosis (TP) in 20 patients with 1 false positive results (FP). In the rest of the 4 patients with negative PET/CT final specific diagnosis (FN) was made only in 2patients (2/4-50%) and in 2 patients no diagnosis was made (TN). The PET/CT had a sensitivity of 90.91%, specificity of 66.67%, positive predictive value (PPV) of95.24% and negative predictive value (NPV-50 %). TP cases included 8 cases of Kochs ( 2 cases of tubercular lymphadenitis, 1 case of tubercular pleural effusion, 1 case of spinal TB, 1 case of genitourinary TB, 1 case of Cutaneous TB, 1 case of splenic TB; 1 case of abdominal TB); 4 cases of focal infections/abscesses ,1 cases of lymphoma, 2 cases of Sarcoidosis , 1 case of Kikuchi Fusimoto, 1 case of aortoarteritis, 1 case of Castleman's disease,1 case of Rosai Dorfman and 1 case of thyroiditis. FN negative cases included one case of meningitis and one case of giant cell arteritis. One false positive case was due to focal hyper metabolism on PET in the caecum.

In comparison, on CECT, probable etiology was suspected in 11 patients, out of which 8 cases were true positive (TP), 3 cases were false positive and 12 were false negative. In 2 patients no diagnosis was made (TN). CECT had a sensitivity of 42.86%,

specificity of 50% positive predictive value (PPV) of 84% and negative predictive value (NPV-14.29 %).

Thus 18F FDG PET/CT improved the sensitivity (90.91% vs 42.86 %), specificity (66.67 % vs 50%), Positive predicate value (95.24% vs 84 %) and Negative predicate value (50 % vs 14.29 %) as compared CECT alone. In addition, FDG PET/CT also helped in deciding the metabolically active site for biopsy/FNAC.

4. DISCUSSION:

PET/CT detects the metabolic changes far ahead of the morphological changes which is delineable by other modalities thus aiding early detection of the lesions which may not be apparent on CT. Hyper metabolism in normal sized diseased lymph nodes can be demonstrated only on PET. One of the splenic tuberculosis cases in our study did not reveal any positive finding on CECT. However, PET showed multiple focal FDG avid lesions in spleen which on histopathology confirmed Tuberculosis. Just for academic interest MR abdomen was performed for this patient. The splenic lesions were not visualized on MR as well. Thus this patient would go undiagnosed for splenic tuberculosis, if PET/CT had not been performed.

Early identification of metabolic changes by PET prior to the morphological changes manifested on CECT could explain the relative low sensitivity of the CECT as compared to PET/CT.

PET/CT can sometimes lead to false positive findings. Bowel uptake is one such area where caution should be exercised while reporting PET. One case of PUO which showed focal hyper metabolism on PETCT was thought to be cecal inflammatory pathology or infection, however on subsequent colonoscopic biopsy, non specific findings were obtained.

In two false negative cases, no significant activity was demonstrated on the PETCT. One patient continued to worsen and developed neck rigidity. The CSF study performed subsequently revealed meningitis. This can be explained as whole body PET has limited imaging role in meningitis due to physiological uptake in that region. Another case of giant cell arteritis did not show any hypermetabolic activity in the temporal arteries. However subsequent biopsy confirmed the giant cell arteritis of the temporal artery. This can be explained by the fact that PETCT is not sensitive for small vessel disease.

4.1 PATHOPHYSIOLOGY

18F-Labeled FDG is a structural analog of 2-deoxyglucose with a half-life of 110 min. Three mechanisms of transport have been described for the uptake of glucose into the tissues- Passive Diffusion, Na-dependent glucose transporter (GLUT), facultative GLUT-1 through GLUT-13 transporter [12].

18F-FDG is phosphorylated in the human cells to 29-FDG-6 phosphate by the hexokinase enzyme. 29-FDG-6 phosphate does not participate in the glycolytic

pathway or the pentose-phosphate shunt and gets trapped in the infective / inflammatory cells as it cannot be metabolized, nor can it diffuse back into the extracellular space [figurel]. Over expression of distinct facultative GLUT isotypes and overproduction of glycolytic enzymes in tumoral cells, explains the accumulation of 18F-FDG [13]. Regardless of whether the tissue of origin expresses these enzymes, in malignant cells, there is over expression of type II hexokinase and to a lesser extent type I hexokinase [14].

Increased FDG uptake is present in all activated leukocytes (granulocytes, monocytes as well as lymphocytes), which enables imaging of acute and chronic inflammatory processes [5].

4.2 ETIOLOGIES 4.2.1 Infectious conditions

18F-FDG PET has about 90% sensitivity in diagnosing abdominopelvic abscesses, active tuberculosis, atypical pneumonias, renal abscess, focal nephritis [Figure 2a and 2b], bacterial colitis, diverticulitis, and infected vascular grafts [15]. FDG PET/CT picks up early infectious processes even when other diagnostic procedures are normal. Tuberculosis is the most common infectious disease that causes PUO in developing countries [16] [Figure 3, 4, 5].

Infective endocarditis is another entity that can cause PUO. There are certain pitfalls of echocardiography. As small vegetations in patient's prosthetic heart valves may be obscured by the intense echoes produced by the prosthesis, it may be difficult for them to be detected on echocardiography. In spite of the normal myocardial FDG uptake, FDG PET accurately helps identify sites of infective endocarditis in majority of the patients [17].

Pericarditis: inflammatory process that can result from localized or systemic diseases, including infection, connective tissue disorders, and uremia show mild FDG uptake. Pericardial thickening, pericardial enhancement, minimal to mild FDG uptake, and pericardial fluid are some of the imaging features [Figure 6] on FDG PET/CT [18].

Soft tissue infections: PET easily locates the sites of the soft tissue infections. In musculoskeletal system FDG PET/CT can demonstrate new sites of infection and can subsequently guide surgical management [19]. Asymptomatic chronic osteomyelitis, especially of the central skeleton can present as PUO. Accuracy is said to be about 97% for peripheral skeleton and 93% for axial skeleton [15].

Orthopedic prosthesis can cause fever for prolonged period time when infected. Though sensitivity is high, specificity is less as differentiation between abrasion-induced inflammation and bacterial inflammation is not always possible by 18F-FDG PET. A negative PET in a case of PUO in suspected prosthesis infection eliminates the need for further investigations or revision surgery. However, a positive result does not differentiate inflammation and infection. Aseptic loosening and infection of a prosthetic joint are both accompanied by an inflammatory response, hence are difficult to differentiate on PET/CT [17, 20]. FDG uptake can be increased in inflammatory arthritis, in acute fractures, and in normally healing bone up to 4 months after surgery. These factors are to be excluded before deciding the cause of increased

FDG uptake [17]. The role of FDG PET has not been established in diagnosis of meningitis as cause of fever due to physiological uptake in that region [21].

4.2.2 Noninfectious Inflammatory Diseases

Autoimmune and connective tissue diseases, vasculitis syndromes, granulomatous disorders (sarcoidosis, Crohn's disease), subacute thyroiditis and "miscellaneous" diseases, constitute about 15%-30% of all causes of PUO [15].

Vasculitis is pathologically characterized by blood vessel inflammation with leukocytic infiltration in the vessel wall and reactive damage to mural structures and surrounding tissues [15]. Takayasu's arteritis mainly affects young females between the ages of 10 and 40 years. The disease is most prevalent in Asian populations [18]. Large-vessel vasculitis, especially giant cell arteritis (GCA) is mainly seen in older patients, older than 50 years more common in Scandinavian and North American countries [18]. PET plays a pivotal role in identifying vasculitis in patients who are referred for whole-body imaging for constitutional symptoms and fever of unknown origin, before anatomic changes are identifiable at CT or MR imaging. Circumferential increased metabolic activity in the vessel walls is considered diagnostic of active disease on FDG /PET [18]. Wall thickening, wall enhancement, and alternating focal areas of luminal narrowing and dilatation are characteristic features on CT and MR[18]. A homogeneous smooth linear and long segmental uptake in the thoracic aorta and its main branches [Figure7], with higher intensity compared to the liver, is considered to be a characteristic pattern of giant cell arteritis (GCA) [20]. Similar findings are also seen in Takayasu's arteritis.

A positive hypoechogenic halo on duplex sonography correlates well with FDG PET in larger arteries, but not in arteries less than 4mm [15,21], since the spatial resolution of F18-FDG PET is of 4-6 mm. Thus the usefulness is limited to medium and small vessel inflammation [19].

Also FDG/PET helps in assessment of extent of disease in the whole body. Thus whole-body 18F-FDG/ PET can be used as the investigation of choice if vasculitis of the large arteries is suspected, because the chance of a positive finding may be higher with PETCT than with MRI [15, 21]. Disease activity and monitoring response to therapy is another area where FDG PET/CT appears to be reliable non invasive investigation. Early treatment due to early diagnosis would help prevent life threatening complications of these conditions [15, 21].

Vasculitis of medium and small vessels like Churg-Strauss syndrome, Wegener's disease, and polyarteritis nodosa are less studied and FDG PET would help in detection only when larger vessels are involved [15,21].

Sarcoidosis, a multisystem granulomatous disease of unknown etiology, is another cause of PUO [Figure 8]. Elevated FDG uptake in mediastinal and hilar nodes, without other radiological findings may be seen on FDG PET/CT. Visualization of the so-called "Lambda sign" is a helpful clue [16]. Again FDG uptake and FDG/PET could be used for monitoring response to treatment as disease activity correlates with the degree of uptake, which is an additional advantage of FDG PET/CT [7]. However FDG/ PET

cannot distinguish sarcoidosis from other causes of lymphadenopathy with uptake in the same locations [15].

Crohn's disease: 18F-FDG/ PET had a sensitivity of 85.4% and a specificity of nearly 90% in detecting foci of uptake in atypical Crohn's disease [15].

Subacute thyroiditis: is a quite common noninfectious, inflammatory cause of PUO [15].Inflammation of the thyroid tissue is accompanied by decreased glandular iodine content (decreased attenuation on CT). Diffusely increased FDG avidity [Figure 9] on PET supports a diagnosis of thyroid inflammation [16].

Kikuchi-Fujimoto disease (KFD): is histiocytic necrotizing lymphadenitis, a rare and benign condition described in young women, characterized by cervical lymphadenopathy and fever. This condition is often mistaken for more serious conditions [22]. An immune response of the T cells and histiocytes to an inciting agent is said to be behind pathogenesis. Fever with lymphadenopathy is the usual presentation [axillary, mediastinal, iliac, intraparotid, retrocrural, peripancreatic and celiac nodes], cervical lymphadenopathy being most common [Figure 10]. Other causes of generalized lymphadenopathy including tuberculosis (TB), lymphoma, HIV-AIDS, toxoplasmosis, secondary syphilis, Lymphogranuloma venereum and Kawasaki disease are always to be considered in the differential diagnosis [23]. Common findings on CT are diffusely enhancing lymph nodes. Many cases have been wrongly diagnosed as tuberculosis or lymphoma. Hence definitive diagnosis is only by biopsy. PET/ CT helps in locating site of accessible, active node for biopsy. Though considered as a self-limiting disease and managed supportively, it needs follow up as these patients tend to develop autoimmune diseases like systemic lupus erythematosis [22].

Rosai-Dorfman Disease [Multicentric sinus histiocytosis] is a rare, benign granulomatous disease that typically presents with massive cervical lymphadenopathy [Figure 11]. The disease can be extranodal in 43% and it usually includes skin, soft tissues, respiratory system, genitor urinary system, bones, central nervous system (CNS), orbit, thyroid and breast. CNS and renal involvement is uncommon. Fever, leukocytosis, increased erythrocyte sedimentation rate and hypergammaglobulinemia are the common presentations. The final diagnosis is by histologic examination, which typically shows intrasinus histiocytic proliferation with cells showing lymphocytophagocytosis. Rosai-Dorfman disease should be considered in the differential diagnosis of granulomatous infection, pseudo granulomatous lesion and malignancy [24].

Castleman disease is also known as angiofollicular lymph node hyperplasia or giant lymph node hyperplasia. Castleman disease is one of the causes of non-neoplastic lymphadenopathy. It commonly involves lymphatic tissues. Lungs, larynx, parotid glands, pancreas, meninges, and muscles are the sites of extra lymphatic involvement. Plasma cell and hyaline types are the two varieties described; the former most commonly presenting in the form of multicentric disease. Hyaline vascular disease is more common and is unicentric in 90% of the cases. It usually manifests as an asymptomatic mass lesion with a benign course. Plasma cell castleman disease is commonly associated with fever, night sweats and malaise; hematologic and

immunologic abnormalities such as anemia, thrombocytopenia, hyperglobulinemia and splenomegaly.

In patients with known primary malignancy, F18 FDG PET/CT will help differentiating metastatic node from benign activity as in castleman disease. Usually, the difference between the glycolytic activity of the involved nodes and the primary tumor suggests the possibility of non metastatic nodal disease, including Castleman disease.The systemic forms of the disease especially are associated with an increased risk for neoplasms like Kaposi sarcoma and follicular dendritic cell tumors, non-Hodgkin lymphoma, Hodgkin disease and plasmacytoma [25]. Deep vein thrombosis as the cause of PUO is reported in 2% to 6% of patients [20].

4.2.3 Malignancies:

FDG PET/CT helps in detection of occult neoplasms. Other than lymphomas [ Figure12] , Colorectal cancer, pancreatic cancer, and soft-tissue sarcomas have also been described as causing PUO and are commonly detected by FDG PET/CT. Malignancies presenting with PUO might represent an early stage of the disease process [16]. Hence early detection of malignancy may help in initiation of treatment with a curative intent. In a study by Sorensen et al, patients with PUO had higher risk for hematologic malignant diseases; sarcoma; and cancers of the liver, brain, kidney, colon, and pancreas [15, 26]. FDG PET scan, may not be able to detect a limited variety of systemic (non-focal) diseases like leukemia which may be the cause of PUO.

In HIV infected patients: FDG PET/CT is having a major role in AIDS patients with central nervous system involvement. Both toxoplasmosis and lymphoma are not infrequent CNS complications of AIDS.

CNS lymphoma is highly metabolically active, whereas toxoplasmosis is not. Hence FDG PET/CT plays a major role here when CT and MR fail to differentiate the two. Quantitative assessment has shown that the standardized uptake values of toxoplasmosis are significantly lower than those of lymphoma, with virtually no overlap between the uptake values of the two conditions, thus asserting the importance of FDG PET [17].

Special imaging considerations have to be kept in mind in FDG PET imaging in these patients [a] Inflammatory foci in the lungs are common in immunosuppresed patients and are a common cause of FDG hyper metabolism [b] There is a clear association between the pattern of lymphoid tissue activation and HIV progression. FDG uptake increases in acute disease, in the head and neck. Mid-stages show hyper metabolism in cervical, axillary and inguinal lymph nodes. In later stages FDG accumulation is seen in the colon along with mesenteric and ileocecal lymph nodes [27]. Awareness of these patterns are important [c] Also, splenic FDG uptake greater than hepatic FDG uptake is observed in HIV patients. Thus lymph nodes found to be positive by FDG/PET alone may require further study to differentiate malignancy from inflammatory processes, particularly with high viral loads and advanced stage disease [27].

In a study by Becerra Nakayo et al [28] on cost effectiveness of FDG PET/CT in PUO, €5471 per patient would have been saved if PET/CT had been performed earlier in the

work up. They concluded that PET/CT study could not only be cost-effective in the PUO process if used at an early stage, by establishing an early diagnosis, but also would reduce hospitalization days due to diagnostic purposes and the repetition of unnecessary tests.

5. CONCLUSION

FDG PET/CT is better than CECT in making a specific diagnosis in the evaluation of PUO. It also helps in deciding the best possible site for biopsy. From our study we can infer that 18F-FDG PET/CT may be used as an initial noninvasive diagnostic tool for assessment of patients with PUO.

REFERNCES

[1] PETERSDORF RG, BEESON PB. Fever of unexplained origin: report on 100 cases. Medicine. 1961 Feb 1;40(1):1-30.

[2] Durack DT, Street AC. Fever of unknown origin--reexamined and redefined. Current clinical topics in infectious diseases. 1991; 11:35.

[3] Roth AR, Basello GM. Approach to the adult patient with fever of unknown origin. American family physician. 2003 Dec;68(11):2223-8.

[4] Hersch EC, Oh RC. Prolonged febrile illness and fever of unknown origin in adults. American family physician. 2014 Jul;90(2):91-6.

[5] Rao A, Mallarajapatna G, Mallareddy S, Shankar M, Patil A, Shivakumar S. ECR 2016 / C-0020 / Beyond Lymphoma And Tuberculosis: Role of imaging By 18F-FDG PET/CT In Pyrexia Of Unknown Origin. - EPOS™ [Internet]. Posterng.netkey.at. 2016. Available from: http://posterng.netkey.at/esr/viewing/index.php?module=viewing_poster&pi=131464

[6] Singh N, Kumar R, Malhotra A, Bhalla AS, Kumar U, Sood R. Diagnostic utility of fluorodeoxyglucose positron emission tomography/computed tomography in pyrexia of unknown origin. Indian journal of nuclear medicine: IJNM: the official journal of the Society of Nuclear Medicine, India. 2015 Jul;30(3):204.

[7] Lorenzen J, Buchert R, Bohuslavizki KH. Value of FDG PET in patients with fever of unknown origin. Nuclear medicine communications. 2001 Jul 1;22(7):779-83.

[8] Bleeker-Rovers CP, de Kleijn EM, Corstens FH, van der Meer JW, Oyen WJ. Clinical value of FDG PET in patients with fever of unknown origin and patients suspected of focal infection or inflammation. European journal of nuclear medicine and molecular imaging. 2004 Jan 1;31(1):29-37.

[9] Jaruskova M, Belohlavek O. Role of FDG-PET and PET/CT in the diagnosis of prolonged febrile states. European journal of nuclear medicine and molecular imaging. 2006 Aug 1;33(8):913-8.

[10] Federici L, Blondet C, Imperiale A, Sibilia J, Pasquali JL, Pflumio F, Goichot B, Blaison G, Weber JC, Christmann D, Constantinesco A. Value of 18F-FDG-PET/CT in patients with fever of unknown origin and unexplained prolonged inflammatory

syndrome: a single centre analysis experience. International journal of clinical practice. 2010 Jan 1;64(1):55-60.

[11] Buysschaert I, Vanderschueren S, Blockmans D, Mortelmans L, Knockaert D. Contribution of 18 fluoro-deoxyglucose positron emission tomography to the work-up of patients with fever of unknown origin. European journal of internal medicine. 2004 Jun 30; 15(3): 151-6.

[12] Meller J, Sahlmann CO, Scheel AK. 18F-FDG PET and PET/CT in fever of unknown origin. Journal of Nuclear Medicine. 2007 Jan 1;48(1):35-45.

[13] Pauwels EK, Sturm EJ, Bombardieri E, Cleton FJ, Stokkel MP. Positron-emission tomography with [18F] fluorodeoxyglucose. Journal of cancer research and clinical oncology. 2000 Sep 1;126(10):549-59.

[14] Mathupala SP, Rempel A, Pedersen PL. Glucose catabolism in cancer cells. Isolation, sequence, and activity of the promoter for type II hexokinase. Journal of Biological Chemistry. 1995 Jul 14;270(28):16918-25.

[15] Meller J, Sahlmann CO, Scheel AK. 18F-FDG PET and PET/CT in fever of unknown origin. Journal of Nuclear Medicine. 2007 Jan 1;48(1):35-45.

[16] Nazar AH, Naswa N, Sharma P, Soundararajan R, Bal C, Malhotra A, Kumar R. Spectrum of 18F-FDG PET/CT findings in patients presenting with fever of unknown origin. American Journal of Roentgenology. 2012 Jul;199(1):175-85.

[17] Love C, Tomas MB, Tronco GG, Palestro CJ. FDG PET of Infection and Inflammation 1. Radiographics. 2005 Sep;25(5):1357-68.

[18] James OG, Christensen JD, Wong TZ, Borges-Neto S, Koweek LM. Utility of FDG PET/CT in inflammatory cardiovascular disease. Radiographics. 2011 Sep 6;31(5): 1271-86.

[19] London K, Howman-Giles R. FDG PET/CT in the investigation of PUO, inflammatory and infective conditions in children. InSociety of Nuclear Medicine Annual Meeting Abstracts 2011 May 1 (Vol. 52, No. Supplement 1, p. 1398).

[20] Balink H, Verberne HJ, Bennink RJ, van Eck-Smit BL. A rationale for the use of F18-FDG PET/CT in fever and inflammation of unknown origin. International journal of molecular imaging. 2012 Dec 17;2012.

[21] Meller J, Sahlmann CO, Gurocak O, Liersch T, Meller B. FDG-PET in patients with fever of unknown origin: the importance of diagnosing large vessel vasculitis. The Quarterly Journal of Nuclear Medicine and Molecular Imaging. 2009 Feb 1;53(1):51.

[22] Vivekanandarajah A, Krishnarasa B, Hurford M, Gupta S. Kikuchi's disease: a rare cause of fever and lymphadenopathy. Clinical medicine insights. Pathology. 2012;5:7.

[23] Aneja A, Sheikh S. A rare case of multifocal lymphadenopathy in a young male. Oxford Medical Case Reports. 2014 Nov 1;2014(8):141-4..

[24] Chandrashekhara SH, Manjunatha YC, Muzumder S, Bahl A, Das P, Suri V, Thulkar S. Multicentric sinus histicytosis (Rosai-Dorfman disease): Computed tomography, magnetic resonance imaging findings. Indian journal of medical and paediatric oncology: official journal of Indian Society of Medical & Paediatric Oncology. 2011 Jul;32(3):174.

[25] Bonekamp D, Horton KM, Hruban RH, Fishman EK. Castleman disease: the great mimic. Radiographics. 2011 Oct 4;31(6):1793-807.

[26] Sorensen HT, Mellemkj^r L, Skriver MV, Johnsen SP, Norgárd B, Olsen JH, Baron JA. Fever of unknown origin and cancer: a population-based study. The lancet oncology. 2005 Nov 30;6(11):851-5.

[27] Davison JM, Subramaniam RM, Surasi DS, Cooley T, Mercier G, Peller PJ. FDG PET/CT in Patients With HIV. American Journal of Roentgenology. 2011 Aug;197(2):284-94.

[28] Nakayo EB, Vicente AG, Castrejón AS, Narváez JM, Rubio MT, García VP, García JC. Analysis of cost-effectiveness in the diagnosis of fever of unknown origin and the role of 18 F-FDG PET-CT: A proposal of diagnostic algorithm. Revista Española de Medicina Nuclear e Imagen Molecular (English Edition). 2012 Aug 31;31(4): 178-86.

Figure Key

Figure 1: Warburg effect

Figure 2: 2a [Left]: A 43-year-old man with transplant kidney presenting with PUO; Hyper metabolic activity seen in the right renal cortex corresponding to the thickening and focal perinephric stranding on CT Renal abscess 2b[Right] A 43 year old male patient with transplant kidney presenting with PUO: Hyper metabolic focus seen in the upper pole of transplanted kidney suggested possibility of focal nephritis. Patient became symptom free after a course of appropriate antibiotics.

Figure 3.A 72 year old maleadult presenting with PUO. 18F FDG PET/CT showing multiple enlarged FDG avid lymphnodes in the right supraclavicular region and mediastinum [paratracheal, left hilar and subcarinal].Lymph nodal biopsy revealed Tuberculosis. Figure 4A: A 70 year old woman presenting with PUO and back pain. 18F FDG PET/CT showed destructive changes in the L1 vertebral bodies, endplates and the intervening L1-L2 disc with soft tissue in the adjacent pre and paravertebral region. Intense FDG uptake seen in the respective areas. CT guided biopsy of the paravertebral soft tissue revealed tuberculosis. Figure 4B: Ill-defined hypodense areas in the left renal parenchyma with low-grade FDG uptake. Also there is blunting of renal calyces and narrowing of the infundibula, suggestive of probable renal tubercular involvement in the same patient.

Figure 5:A 16 year old female presenting with PUO: This was a unique case with respect to

imaging where all the modalities including USG, CECT, MRI were normal, only PET-CT

showing nodular uptake in the spleen. CT guided PET/CT directed splenic biopsy revealed

tubercular infection of spleen. Thus only PET CT could detect and guide the site of biopsy

when all modalities failed. This highlights the role of PET/CT as functional imaging tool

where lesions were detected when they were morphologically not evident.

Figure 6:A 44 year old female presenting with PUO, CT image showing moderate pericardial

effusion with mild pericardial enhancement. Corresponding PET images and fused PET/CT

images show mild FDG uptake in the pericardium- suggestive of pericarditis.

Figure 7:A 18 year old female presenting with PUO: Mild diffuse FDG uptake seen along the

arch and descending thoracic aorta corresponding to the subtle wall thickening on CT:

Aortoarteritis

Figure 8: A 51 year old lady presenting with PUO. FDG PET/CT showing hypermetabolic mediastinal lymph nodes seen with increased FDG uptake in the pituitary gland and adrenals. Mediastinal lymph node biopsy revealed sarcoidosis

Figure 9: A 54 year old lady with PUO: PETCT showing diffuse FDG uptake in the thyroid Figure 10: A 30 year old male with PUO: Multiple hypermetabolic lymph nodes in the neck, mediastinum and abdomen with mild hepatosplenomegaly and mild diffuse hypermetabolic activity in the spleen. Biopsy from one of the lymph nodes revealed Kikuchi fujimoto'sdisease Figure 11: A 38 year oldfemale presenting with PUO: Right hilar/mediastinal, cervical [not shown] lymphadenopathy with moderate FDG uptake, Mild uptake in right intraparotid lymph node; metabolic activity is seen in the nasopharynx and tonsils: biopsy of the lymphnodes revealed Rosai Dorfman Disease.

Figure 12: A 54 year old female with PUO: Splenomegaly with multiple hypermetabolic intra-splenic lesions, extensive abdominal and mediastinal lymphadenopathy and a discrete

mediastinal lymph node- Splenic biopsy revealed lymphoma. Table legends:

Table 1: showing distribution of cases with histopathological correlation.

Table 2: showing CECT and histopathological correlation.

Table 3: showing 18 F-FDG PET/CT and histopathological correlation.

Table 1

Sr No Age/ Sex Clinical history CECT Diagnosis 18F FDG PET/CT Diagnosis Histopathological Diagnosis

1 48Y/ M PUO with breathlessness Loculated left Pleural effusion (TP) Loculated left pleural effusion with activity in the pleura(TP) Tuberculous effusion

2 72Y/ M PUO with weight loss Indeterminate (FN) Hypermetabolism in mediastinal nodes(TP) Tuberculous lymphadenitis

3 42Y/ M Transplanted kidney with PUO Hypo enhancing area at the upper pole of transplanted kidney (TP) Focal uptake at the upper pole of transplanted kidney(TP) Focal pyelonephritis

4 35Y/ F PUO Indeterminate (TN) Indeterminate(TN) True PUO

5 16Y/ F PUO Indeterminate(FN); No positive findings on CECT multifocal hypermetabolic activity in the spleen(TP) Splenic Koch

6 32Y/ M PUO with cough Mediastinal, Axillary Lymphadenopathy (TP) Hyper metabolism in Mediastinal lymphadenopathy(TP) Sarcoid

7 38Y/ F PUO with Right upper quadrant pain Indeterminate (FN) Focal activity in the gall bladder wall-Cholecystitis (TP) Acute cholecystitis

8 54Y/ F PUO with axillary swelling Mediastinal, Axillary and Abdominal Lymphadenopathy (TP) Hyper metabolism in Mediastinal, Axillary and Abdominal Lymphadenopathy(TP) Lymphoma

9 44Y/ F PUO with breathless ness Indeterminate (FN) Hyper metabolism in pericardium+ Minimal Pleural /Pericardial effusion(TP) Infectious pericarditis

10 51Y/ F PUO with weight loss Hepatosplenomegaly+ Lymphadcnopathy(FP) Hyper metabolism in pituitary and adrenals (TP) Sarcoidosis

11 30Y/ M PUO Indeterminate(FN) Hyper metabolism in neck, mediastinum + abdomen (TP) Kikuchi Fusimoto

12 18Y/ F PUO with malaise Indeterminate(FN) Hyper metabolism in ascending and arch of aorta (TP) Aortoarteritis

13 54Y/ M PUO with malaise Lymphadenopathy (TP) Hypermetabolism in inguinal lymph nodes(TP) Castleman's disease

14 38Y/ F PUO Indeterminate(FN) Hypermetabolism in right hilar/mediastinal, cervical + right intraparotid nodes (TP) Rosai Dorfman

15 42Y/ M PUO with malaise Indeterminate (TN) Indeterminate(TN) True PUO

16 36Y/ F PUO with weight loss Indeterminate (FN) Hypermetabolism in ear lobule(TP) Cutaneous Koch

17 70Y/ F PUO with back pain Reduced intervertebral disc height with vertebral collapse (TP) Hypermetabolism in LI vertebra and L1-L2 Intervertebral disc(TP) Spinal Koch

18 62Y/ M PUO with cough and breathlessness Mediastinal, Axillary and Abdominal Lymphadenopathy (TP) Hypermetabolism in Mediastinal > Axillary + Abdominal Lymphadenopathy (TP) Tuberculous lymphadenitis

19 47Y/ M PUO with left flank pain Focal pyelonephritis in left upper pole (TP) Focal Hypermetabolism in left renal upper pole (TP) Pyelonephritis

20 54Y/ M PUO with weight loss Indeterminate(FN) Ileocecal + Omental + Lymph nodal hyperactivity (TP) Abdominal Koch

21 43Y/ M PUO in a patient with renal transplant Renal abscess(TP) Renal abscess with omental activity (TP) Tubercular renal abscess

22 54Y/ F PUO Indeterminate (FN) Intense hypermetabolic activity in thyroid gland (TP) Thyroiditis

23 32Y/ M PUO with neck pain Indeterminate (FN) Indeterminate (FN) Meningitis

24 52Y/ F PUO Indeterminate (FN) Indeterminate (FN) Giant cell arteritis

25 43Y/ F PUO with malaise Minimal ceacal wall thickening (FP) Focal hyper-metabolism in caecum (FP) True PUO

Table 2

Histopathology Positive Negative

CECT Positive 9 2

CECT Negative 12 2

Table 3

Histopathology POSITIVE NEGATIVE

18F FDG PET/CT. Positive 20 1

18F FDG PET/CT. Negative 2 2

WELL DIFFERENTIATED TISSUE CELLS

PROLIFERATING TUMORS CELLS

GLUCOSE GLUCOSE

+02 PYRUVATE^ PYRUVATE

LACTATE

OXIDATIVE PHOSPHORYLATION

k LACTATE

LACTATE ANEROBIC GLCOLYSIS

AEROBIC GLCOLYSIS

Figure liWARBURG EFFECT

£). / ,/v) v /v