564 5. Lunca S, Romedea N. Actinomycosis of the appendix. Case report. Rev Med Chir Soc Med Nat Iasi 2004;108:640—3. 6. Habib E, Aoura T, Mekkaoui M, Elhadad A. Actinomycosis of the appendix. Report of two cases. Rev Med Interne 2002;23:638—41. 7. Davies M, Keddie NC. Abdominal actinomycosis. Br J Surg 1973;60:18—22. 8. Meyer P, Nwariaku O, McClelland RN, Gibbons D, Leach F, Sagalowsky AI, et al. Rare presentation of actinomycosis as an abdominal mass: report of a case. Dis Colon Rectum 2000;43: 872—5. 9. Dayan K, Neufeld D, Zissin R, Bernheim J, Paran H, Schwartz I, et al. Actinomycosis of the large bowel: unusual presentations and their surgical treatment. Eur J Surg 1996;162:657—60. 10. Goldwag S, Abbit PL, Watts B. Case reports: percutaneous drainage of periappendiceal actinomycosis. Clin Radiol 1991;44: 422—4. 11. Maradeix S, Scrivener Y, Grosshans E, Sabatier X, Riegel P, Cribier B. Actinomycosis of the buttock. Ann Dermatol Venereol 2005;132:462—5. 12. Mumme T, Peiper C, Biesterfeld S, Schumpelick V. Acute retrocecal appendicitis caused by an Actinomyces israelii mixed infection. Zentralbl Chir 2001;126:632—4.
Lymphomatous pericardial effusion positive for Mycobacterium tuberculosis by PCR analysis We report the case of a female patient with lymphomatous pericardial effusion, which was found to be positive for Mycobacterium tuberculosis by PCR analysis. A 64-year-old woman presented to the hospital complaining of left hemithorax discomfort and vomiting. Lung auscultation revealed diminished breath sounds at both bases, heart auscultation revealed low cardiac valve tones. The patient was afebrile, while liver, spleen, and peripheral lymph nodes were not palpable. Examination of the skin revealed a small number of abdominal subcutaneous nodules. An echocardiograph revealed pericardial effusion and the patient was admitted. Pericardiocentesis produced 500 ml of fluid, which was sent for analysis and culture as well as PCR analysis for M. tuberculosis. One of the subcutaneous nodules was removed and sent for histological analysis. On laboratory analysis, peripheral blood values were: hematocrit 34.1%, mean corpuscular volume 91.7 fl, hemoglobin 11.1 g/dl, white blood cells (WBC) 7.4 109/l (neutrophils 6.0 109/l, lymphocytes 0.755 109/l, eosinophils 0.1 109/l, monocytes 0.5 109/l), platelets 318 109/l, erythrocyte sedimentation rate 48 mm/h, glucose 135 mg/ dl, creatinine 0.5 mg/dl, total bilirubin 1.12 mg/dl, lactate dehydrogenase (LDH) 918 U/l, C-reactive protein 3.63 mg/ dl, total protein 5.4 g/dl, albumin 3.2 g/dl. Pericardial fluid values were: red blood cells 800 109/l, WBC 50 109/l (neutrophils 4%, lymphocytes 2%, undetermined 94%), glucose 7 mg/dl, LDH 12 000 U/l, albumin 4.6 g/dl, cholesterol 143 mg/dl. A tuberculin skin test was negative. Autoimmune and tumor markers were normal. Viral serology (hepatitis B, cytomegalovirus, Epstein—Barr virus, HIV) was negative. Pericardial fluid demonstrated negative Ziehl—Nielsen staining. Cultures of pericardial fluid were negative for microorganisms including M. tuberculosis. A sample was sent for
Correspondence C. Nissotakis G.H. Sakorafas* T. Koureta K. Revelos G. Kassaras G. Peros Department of Surgery, 251 Hellenic Air Force Hospital, Athens, Greece *Corresponding author. Tel.: +30 (210) 74 87 318; fax: +30 (210) 74 87 192 E-mail address:
[email protected] (G.H. Sakorafas) Corresponding Editor: James Muller, Pietermaritzburg, South Africa 16 July 2007 doi:10.1016/j.ijid.2007.12.015
cytological examination. A chest computed tomography scan demonstrated the presence of pericardial effusion and minor pleuritic effusion, withoutdenopathy. Pericardiocentesis was performed twice more, and all three samples of fluid presented positive for M. tuberculosis by PCR. The Amplicor PCR assay was used (Roche Molecular Systems), with which the optical density at 450 nm was >0.400 in all three specimens. Anti-tuberculosis treatment was initiated (streptomycin, rifampin, isoniazid, and ethambutol). At this point we must stress that the patient had no prior history of tuberculous infection. The skin biopsy demonstrated infiltration with atypical CD20(+) lymphocytes, indicating high malignancy diffuse Bcell lymphoma. Cytology of the pericardial effusion indicated the same neoplastic cells. The patient, parallel to anti-tuberculosis therapy, was started on a CHOP (cyclophosphamide, doxorubicin, vincristine, prednisolone) chemotherapeutic regimen. After six treatment cycles the patient was in remission and without pericardial effusion. Pericardial effusion due to non-Hodgkin lymphoma (NHL) is uncommon, presenting in less than 1% of cases.1 Moreover, tuberculous pericarditis is rare, occurring in 1% of cases.2 PCR in pericardial effusion may provide the diagnosis of pericardial tuberculosis when other clinical indicators are absent and cultures are negative.3—5 The sensitivity and specificity of PCR-based assays for M. tuberculosis are reported to be excellent at 90—98% and >98%, respectively, although sensitivity for specimens without acid-fast bacilli seen on direct microscopic examination may be as low as 46%.6,7 In the case presented here, the pericardial effusion proved positive for M. tuberculosis using the PCR method on three separate occasions, and this was considered adequate to initiate anti-tuberculosis therapy, despite the negative culture. The presence of M. tuberculosis in lymphomatous pericardial effusion in the case presented herein was probably due to lymphoma-related immunosuppression, a condition that is documented.8 Lymphomatous pericardial effusion
565
Correspondence should consistently be analyzed for M. tuberculosis and treated accordingly. Conflict of interest: No conflict of interest to declare.
References 1. Das DK, Gupta SK, Ayyagari S, Bambery PK, Datta BN, Datta U. Pleural effusions in non-Hodgkin’s lymphoma. A cytomorphologic, cytochemical and immunologic study. Acta Cytol 1987;31:119— 24. 2. Kim HY, Song KS, Goo JM, Lee JS, Lee KS, Lim TH. Thoracic sequelae and complications of tuberculosis. Radiographics 2001;21:839—58. discussion 859—60. 3. Rana BS, Jones RA, Simpson IA. Recurrent pericardial effusion: the value of polymerase chain reaction in the diagnosis of tuberculosis. Heart 1999;82:246—7. 4. Tzoanopoulos D, Stakos D, Hatseras D, Ritis K, Kartalis G. Detection of Mycobacterium tuberculosis complex DNA in pericardial fluid, bone marrow and peripheral blood in a patient with pericardial tuberculosis. A case report. Neth J Med 2001;59: 177—80. 5. Levy PY, Fournier PE, Charrel R, Metras D, Habib G, Raoult D. Molecular analysis of pericardial fluid: a 7-year experience. Eur Heart J 2006;27:1942—6. 6. Louie M, Louie L, Simor AE. The role of DNA amplification technology in the diagnosis of infectious diseases. CMAJ 2000;163: 301—9.
Causes for low positive predictive values of CD4 counts for antiretroviral treatment failure Chaiwarith et al. made the observation that compared to HIV viral load, CD4 count measurements showed a sensitivity of only 13.3% for detecting antiretroviral failure.1 The authors also noted a low positive predictive value of the CD4 count for antiretroviral treatment failure of 8.6%, which was significantly lower than clinical criteria. This is an important finding indicating that there may be causes for low CD4 counts in HIV patients on antiretroviral treatment different from HIVinduced immune suppression. The causes of low CD4 counts without virological failure require further analysis. The decision to discontinue prophylaxis against opportunistic infections for patients on antiretroviral treatment depends on the CD4 count. Malnutrition can suppress CD4 counts with preserved immune function.2 Other causes of lymphopenia may be the antiretroviral drugs themselves: zidovudine exposure in uninfected infants has been associated with low CD4 counts.3 Patients on tenofovir and didanosine with complete suppression of viral replication have previously been shown to have a paradoxical CD4+ decline, attributed to an imbalance of adenosine metabolites in this cell population.4 In future studies, adjustment for body mass index and side effects of antiretroviral drugs may improve the use of CD4 counts as a tool for monitoring immune function and the success of antiretroviral treatment. Conflict of interest: No conflict of interest to declare.
7. Singh UB, Bhanu NV, Suresh VN, Arora J, Rana T, Seth P. Utility of polymerase chain reaction in diagnosis of tuberculosis from samples of bone marrow aspirate. Am J Trop Med Hyg 2006;75:960—3. 8. Davis SD, Yankelevitz DF, Williams T, Henschke CI. Pulmonary tuberculosis in immunocompromised hosts: epidemiological, clinical, and radiological assessment. Semin Roentgenol 1993;28: 119—30.
Fotios V. Michelis* Costas L. Petrogiannopoulos Kostas X. Papamichael Nikolaos D. Kostakos Eusevia Z. Kiagia Achilleas Deliousis Antonios K. Zacharof 2nd Department of Internal Medicine, Hellenic Red Cross Hospital, Athanasaki 1, Ampelokipi, Athens, Greece *Corresponding author E-mail address:
[email protected] (F.V. Michelis) Corresponding Editor: J. Peter Donnelly, Nijmegen, The Netherlands 3 May 2007 doi:10.1016/j.ijid.2008.02.001
References 1. Chaiwarith R, Wachirakaphan C, Kotarathititum W, Praparatanaphan J, Sirisanthana T, Supparatpinyo K. Sensitivity and specificity of using CD4+ measurement and clinical evaluation to determine antiretroviral treatment failure in Thailand. Int J Infect Dis 2007;11:413—6. 2. Eisenhut M. Malnutrition and variability in CD4+ cell counts in African populations. J Infect Dis 2007;195:1550. 3. Bunders M, Thorne C, Newell ML, European Collaborative Study. Maternal and infant factors and lymphocyte, CD4 and CD8 cell counts in uninfected children of HIV-1-infected mothers. AIDS 2005;19:1071—9. 4. Barrios A, Rendon A, Negredo E, Barreiro P, Garcia-Benayas T, Labarga P, et al. Paradoxical CD4+ T-cell decline in HIV-infected patients with complete virus suppression taking tenofovir and didanosine. AIDS 2005;19:569—75.
Michael Eisenhut Luton and Dunstable Hospital NHS Foundation Trust, Lewsey Road, Luton, LU40DZ, UK E-mail address:
[email protected] Corresponding Editor: William Cameron, Ottawa, Canada 2 December 2007 doi:10.1016/j.ijid.2008.02.002
