A. Mert

Fever of unknown origin: a review of 20 patients with adult-onset Still's disease.

Abstract In this study we aimed to investigate the findings in patients with adult-onset Stills disease (AOSD) admitted with fever of unknown origin (FUO) during the last 18 years in our unit, in order to discover the ratio of such patients to all patients with FUO during the same period, and to determine the clinical features of AOSD in FUO. The number and the aetiologies of the patients with FUO diagnosed between 1984 and 2001, and the clinical features of those with AOSD, were taken from the patient files. The diagnosis of AOSD was reanalysed according to the diagnostic criteria of Cush et al. [11]. The presumed diagnoses before a diagnosis of AOSD was established were also noted. The χ2 and Fishers exact tests were used for statistical analysis. We studied 130 patients with a diagnosis of FUO, 36 (28%) of whom had collagen vascular diseases. Of these 36 patients, 20 (56%, 12 female, 8 male, mean age 34 years, range 16–65) had AOSD. Clinical and laboratory findings were as follows: fever (100%), arthralgia (90%), rash (85%), sore throat (75%), arthritis (65%), myalgia (60%), splenomegaly (40%), hepatomegaly (25%), lymphadenopathy (15%), anaemia (65%), neutrophilic leukocytosis (90%), increased erythrocyte sedimentation rate (100%), elevated transaminase levels (65%), a negative RF (100%), and a negative FANA (80%). Antibiotics had been prescribed in 18 (90%) of cases. The presumed infectious diagnoses were streptococcal tonsillitis/pharyngitis (50%), infective endocarditis (four patients), sepsis (two patients) and acute bacterial meningitis (two patients). The presumed non-infectious diagnoses were acute rheumatic fever (three patients), seronegative rheumatoid arthritis (two patients) and polymyositis (two patients). Sixteen patients were followed for a mean duration of 30 months (range 2–59). A remission was obtained with indomethacin in three cases (19%), and with prednisolone in the remainder. Relapse was detected in three cases (19%). AOSD is one of the most frequent aetiologies of FUO. During the diagnostic course of a patient with FUO, a maculopapular rash and/or arthralgia and/or sore throat should raise the suspicion of AOSD. Because the disease has heterogeneous clinical findings, certain bacterial infections (e.g. streptococcal pharyngitis and sepsis) are generally considered and the prescribing of antibiotics is common.

Fever of unknown origin in Turkey

Abstract.Background:The etiology of fever of unknown origin (FUO) includes primarily infectious, collagen-vascular and neoplastic diseases. The distribution of the disorders causing FUO may differ according to the geographic area and the socioeconomical status of the country. Moreover, the developments in radiographic and microbiologic methods have changed the spectrum of diseases causing FUO.Materials and Methods:We reviewed 117 cases that fulfilled the criteria of FUO followed in our department during the period 1984 to 2001.Results:The etiology of FUO was infectious diseases in 34% of the patients, collagen-vascular diseases in 23%, neoplasms in 19% and miscellaneous diseases in 10%. In 14% of the cases the etiology could not be found. The three leading diseases were tuberculosis (24%), lymphomas (19%) and adult-onset Still’s disease (11%). Tuberculosis was found to be a more common cause of FUO than reported in studies in developed countries. Invasive procedures helped to establish the diagnosis in 50 out of 92 patients (43%). As a final diagnostic procedure, laparotomy aided the establishment of a diagnosis in 15 out of 20 patients (75%).Conclusion:Although the relative rate of infectious disease as etiologic category is less commonly encountered, infectious disease, especially tuberculosis, remains a common cause of FUO. Although several diseases may lead to FUO, lymphomas, adult-onset Still’s disease and particularly tuberculosis should be considered in the differential diagnosis of a patient admitted with FUO.

Tuberculous lymphadenopathy in adults: a review of 35 cases.

Abstract We retrospectively reviewed clinical, diagnostic, therapeutic and prognostic features of 35 patients (25 female, 10 male, mean age: 33 years, range: 16–70) with tuberculous lymphadenopathy (TB LAP) which had been followed since 1980. The diagnosis was established by tissue sampling in 32 cases (caseating granulomatous adenitis in 89%) or presence of acid-fast bacilli (AFB) in the aspirate in 2 cases and in the drainage in 1 case. Paraffin-embedded granulomatous tissues were stained by Ehrlich-Ziehl-Neelsen (EZN) and also Mycobacterium tuberculosis DNA was studied by polymerase chain reaction (PCR) (n = 21). The patients were admitted with enlarging LAP (34%), draining LAP (9%), and both systemic complaints and enlarging LAP (57%). Cervical lymph nodes were the most frequently involved site (77%). Pathologic findings on chest X-ray were seen in 23%. Erythrocyte sedimentation rate (ESR) was higher than 100 mm/hour in 25% and associated with systemic complaints. Tuberculin skin test was positive in 91%. AFB could not be seen in any granulomatous tissue (n: 21), but PCR study was positive in 33% (7/21). All patients were given anti-TB treatment (INH, RMP, EMB and/or PZA). Surgical excision of draining LAP with surrounding inflammatory tissues in addition to the medical treatment was needed in 2 cases. Clinical improvement was obtained within 3 months of the treatment and ESR returned to normal within 5 months. After completion of the treatment, 22 patients were followed-up; mean duration was 3 years, and none relapsed. In conclusion, a patient with TB LAP generally presents with a few small, painless, cervical lymph nodes, which are slowly enlarging. For exact diagnosis, excisional biopsy for histologic and microbiologic studies is essential. Use of anti-TB drugs is the main therapeutic option.

Is laparotomy necessary in the diagnosis of fever of unknown origin

Summary Fever of unknown origin (FUO) is a diagnostic challenge for the practising physician. Detailed medical history, physical examination, non-invasive laboratory tests, and radiologic examinations compose the first level in the diagnostic approach to the FUO. When a diagnosis cannot be established with these procedures, some invasive diagnostic techniques and finally exploratory laparotomy are performed. Although advanced diagnostic measures and imag-ing-guided less invasive procedures have decreased the need, laparotomy remains as a final diagnostic method for FUO cases. In this study we evaluate the role and importance of laparotomy in the diagnosis of our FUO cases. In 17 out of 126 patients (8 male, 9 female, the median age 35.8 years) hospitalized in our clinic between 1982 and 2002 with the diagnosis of FUO, the diagnosis was established by laparotomy. The diagnosis was made directly in 13 patients, and indirectly (by excluding other diseases) in 2 patients. In several FUO series, the contribution of laparotomy to the diagnosis of FUO was reported as 27-100%. This rate was found to be 88% in the present study. During laparotomy on 17 cases, tissue samples were taken from spleen, liver, intra-abdominal and mesenteric lymph nodes. Pathologic examination of these tissue samples revealed miliary tuberculosis in 4; non-Hodgkin’s lymphoma in 3; Hodgkin’s lymphoma in 3; liver tumour in 1; hairy cell leukemia in 1; peritonitis carcinomatosis in 1. In the patients with miliary tuberculosis, the liver (3) and/or spleen (2), and/or lymph node (3) revealed caseating granulomas. Laparotomy diagnosed 3 of 5 cases whose abdominal ultrasonography and computerized tomography were normal. In conclusion, although advanced diagnostic methods decreased the need for laparotomy in FUO, if non-invasive and invasive diagnostic measures fail, laparotomy may contribute to the diagnosis. The selection of the patient and the timing are important for laparotomy.

Is Family Screening Necessary in Brucellosis

Brucellosis is an endemic disease in Turkey. Simultaneous infections among family members consuming infected dairy products have been reported. The most frequent signs and symptoms are nonspecific, and most human cases remain unrecognized. We aimed to screen family members of index cases with brucellosis.A questionnaire including demographical and epidemiological data was obtained. All cases were tested by slide agglutination tests (Rose Bengal test). Seropositive ones were further tested by tube agglutination tests (Wrigth test). In the index cases, Brucella antibody titers of ≥ 1:160 with and without clinical symptoms and 1:80 with clinical symptoms were considered positive and the household members were enrolled into the study. Twenty-eigth index cases were identified among a total of 110 family members. Among family members, 90 (82%) were seronegative whereas 20 were seropositive. Among seropositive cases, 12 were asymptomatic and 8 were symptomatic. The most frequent symptoms of the index cases were fever, headache and arthralgia. Symptomatic cases were treated and asymptomatic ones followed up without therapy for a period of 6–12 months and none developed brucellosis. In conclusion, family members of the patients with brucellosis are under the increased risk of brucellosis because of a common source. Therefore, it can be considered that family screening may lead to early diagnosis of the disease and to the prevention of the complications.

Comparison of community‐onset healthcare‐associated and hospital‐acquired urinary infections caused by extended‐spectrum beta‐lactamase‐producing Escherichia coli and antimicrobial activities

We aimed to compare community‐onset healthcare‐associated (CO‐HCA) and hospital‐acquired (HA) urinary tract infections (UTIs) caused by extended‐spectrum beta‐lactamase (ESBL)‐producing Escherichia coli in terms of epidemiology, clinical outcomes and antimicrobial activities.

Malaria in Turkey: A review of 33 cases

Aim: To evaluate epidemiologic and clinical features of the patients with malaria followed in our clinic, and to review current status of malaria in our country. Patients and methods: Epidemiologic, clinical, diagnostic, and therapeutic features of 33 patients with malaria (4 female, 29 male, mean age: 28 ± 11 years, range: 15–60) followed in our clinic between 1981 and 2000 were evaluated retrospectively. Malaria data of our country for 1926–2000 were obtained from Health Ministry. Results: Diagnosis was established by thin smears of blood preparations obtained in the febrile period in all cases. Plasmodium vivax was detected in 26 patients (25 domestic and one imported), and P. falciparum in seven (two domestic and five imported). Sixty-one percent of the patients had the prodromal symptoms of the disease and used various antibiotics. All cases demonstrated the typical pattern of fever with chills. Fever (100%), splenomegaly (91%), hepatomegaly (55%), anemia (70%), leukopenia (48%), thrombocytopenia (48%), a rise in sedimentation rate (100%), and abnormalities in hepatic enzymes (30%) were determined in the patients. Chloroquine+primaquine were given to all patients with P. vivax, chloroquin (for three) or mefloquin (for four) alone were given to the patients with P. falciparum. One patient with P. falciparum died soon after admission, all the remaining recovered. Data from Health Ministry revealed that the most common (∼100%) species in our country is P. vivax. Although an eradication program against malaria initiated in 1926 achieved success, it still remains as an important health problem. Conclusion: Every febrile patient with a history of travel to the regions where malaria is endemic (tropical regions for the world, southeast regions for our country) should raise the suspicion of malaria. Every country should fight against malaria and global cooperation is essential.

Abdominal actinomycosis: a report of two cases.

Abstract Actinomyces spp. cause a chronic suppurative, granulomatous disease which is characterized clinically by extensive abscess formation, recurrent draining of sinuses and fistulae and histologically by the presence of the so-called “sulphur granules”. Colonic actinomycosis is a relatively rare infection and its diagnosis is difficult. We report the case of a female patient who was operated on for ovarian cyst and the case of another female patient operated on for a mass in the transverse colon. In both cases the pathology of the excised tissues revealed actinomycosis. Actinomycosis must be considered in the differential diagnosis of patients who present with abdominal pain, fever, leu-cocytosis and intestinal wall thickness and /or abdominal mass.

Role of TTV in acute non-A-E hepatitis in Turkey.

Five hepatotropic viruses are the aetiological agents in the majority of acute viral hepatitis cases: HAV, HBV, HCV, HDV and HEV. Their prevalence in the setting of acute hepatitis changes with geographical region. However, in a group of patients with acute hepatitis, serology for these viruses remains negative. The percentage of acute viral hepatitis due to hepatitis-non-A–E viruses has been reported as 13.9% (1). A new, single-stranded and unenveloped DNA virus, known as TTV (transfusion-transmitted virus) was isolated by Japanese investigators in 1997 (2). The prevalence of TTV has been reported as 10% in the general population, 1–1.9% in blood donors, 25% in patients with chronic liver disease, 19–27% in patients with idiopathic fulminant hepatic failure and 15% in patients with cryptogenic cirrhosis (3–5). In spite of a high prevalence of TTV in different liver diseases, current data suggests that it has a pathogenic role in chronic hepatitis (3). The transmission routes of TTV are parenteral or non-parenteral. The latter has been reported as 4% and 38% in 2 different series (3, 5). There is no satisfactory information about vertical, horizontal and sexual transmission. There is a little information about the aetiological role of TTV in acute viral hepatitis cases in MEDLINE research (from 1997 to November 1999) (6). The aim of this study was therefore to determine causes of acute viral hepatitis in Turkey and to assess the relative importance of TTV in the patients with acute viral hepatitis who are seronegative for hepatitis A–E. A total of 109 consecutive patients with acute hepatitis during January 1996 to October 1998 were included in the study. The patients who had an aminotransferase level elevated at least 10fold were included as having acute hepatitis. They were followed in our infectious disease clinic as inpatients or outpatients. The serum samples of inpatients were drawn on the first day of hospitalization. Their mean age was 24 years (range 11–74 years) and 71 were male. Of the 109 cases of acute viral hepatitis, 88% were due to 1 of the 4 recognized hepatitis virus (A–D). ELISA was used for the detection of HbBsAg, anti-HBc-IgM, anti-HAV-IgM (Sanofi Diagnostics Pasteur, Marnes-la-Coqutte, France), anti-HCV, antiCMV-IgM (Dia Sorin Group, Saluggia, Italy), anti-HEV-IgM (Guiliana Cremascoli Chemical, Segrate, Spain), EBV-IgM (Gull, Bad Hamburg, Germany) and parvovirus B19-IgM (Biotrin, Dublin, Ireland). The percentages of HAV, HBV, HCV and HDV were 34.8%, 49.5%, 2.7% and 0.9%, respectively (Fig. 1). It was planned to investigate the 13 seronegative patients further for detection of rare aetiological agents. Serum samples were available for 8 patients, but not for the remaining 5 outpatients. The sera of 8 patients were tested by ELISA for IgM to EBV, CMV, parvovirus B19, HEV; by PCR for HBV-DNA, HCV-RNA and HEV-RNA. All patients were negative for all of the viruses mentioned. The serum samples of these 8 patients, 1 of whom had fulminant hepatitis, were tested for TTV-DNA. TTV-DNA was detected by PCR with semi-nested primers described by Okamoto et al. (7). DNA was extracted from serum with proteinase K-SDS and phenol-chloroform protocol and then TTV-DNA was amplified by semi-nested PCR with TTV-specific primers derived from open reading frame 1 (ORF) of TTV. Amplification products were run by electrophoresis in 2% agarose gel, stained with ethidium bromide and photographed under ultraviolet light. All sera were also negative for TTV. Approximately 12% of cases of acute viral hepatitis in Turkey are of unknown aetiology. This study suggests that TTV is not an aetiological agent of non-A–E acute viral hepatitis in Turkey. One or more additional agents may be responsible for acute viral hepatitis. Therefore, efforts must be made to search for new hepatitis viruses.

Early Changes of Mannose-Binding Lectin, H-Ficolin, and Procalcitonin in Patients with Febrile Neutropenia: A Prospective Observational Study.

Objective: The significance of mannose-binding lectin (MBL) and H-ficolin deficiency in febrile neutropenic (FN) patients and the correlation of these markers along with consecutive C-reactive protein (CRP) and procalcitonin (PCT) levels during the infectious process are investigated. Materials and Methods: Patients with any hematological malignancies who were defined to have “microbiologically confirmed infection”, “clinically documented infection”, or “fever of unknown origin” were included in this single-center prospective observational study. Serum levels of CRP, PCT, MBL, and H-ficolin were determined on 3 separate occasions: at baseline (between hospital admission and chemotherapy), at the onset of fever, and at the 72nd hour of fever. Results: Forty-six patients (54% male, mean age 41.7 years) with 61 separate episodes of FN were evaluated. Eleven patients (23.9%) had “microbiologically confirmed infection”, 17 (37%) had “clinically documented infection”, and 18 (39.1%) had “fever of unknown origin”. Fourteen (30.4%) patients had low (<500 ng/mL) initial MBL levels and 7 (15.21%) had low (<12,000 ng/mL) H-ficolin levels. Baseline MBL and H-ficolin levels did not significantly change on the first and third days of fever (p=0.076). Gram-negative bacteremia more frequently occurred in those with low initial MBL levels (p=0.006). PCT levels were significantly higher in those with microbiologically documented infections. Mean and median PCT levels were significantly higher in cases with bacteremia. There was no significant difference between hemoculture-positive and-negative patients in terms of CRP levels. Conclusion: Monitoring serum H-ficolin levels was shown to be of no benefit in terms of predicting severe infection. Low baseline MBL levels were correlated with high risk of gram-negative bacteremia; however, no significant correlation was shown in the follow-up. Close monitoring of PCT levels is warranted to provide more accurate and specific data while monitoring cases of bacteremia.