Letters to the Editor
Io8
Oesophageal stricture complicating cytomegalovirus ulceration in a patient with AIDS
Accepted for publication 27 November I99I Sir, Cytomegalovirus (CMV) infection is a well-recognised cause of painful oesophageal ulceration in patients with H I V infection. We report a case in which an oesophageal ulcer was complicated by the development of a stricture which was successfully treated. A z3-year-old HIV-positive homosexual man presented with retrosternal pain on swallowing. H e had been treated for Pneumocytis carinii pneumonia 2 years previously. T w o months before admission he had developed oral candidiasis, which responded to oral fluconazole. Despite an increased dose offluconazole he continued to experience pain. Endoscopy demonstrated a typical large mid-oesophageal ulcer, and biopsies showed numerous C M V inclusions. T r e a t m e n t was started with Iv ganciclovir (io m g / k g / 2 4 h via a Hickman line, and parenteral diamorphine was required for pain control. After 25 days there had been some improvement in his symptoms, but neutropenia then developed and ganciclovir was stopped. Another endoscopy showed a partly healed ulcer, with stricturing of the mid-oesophagus, and upper oesophageal candidiasis. T h e dose of fiuconazole was again increased, and he was commenced on phosphonoformate 6o m g / k g 8 hourly for 3 weeks. Following this his pain disappeared, but he developed dysphagia for solid food. A further endoscopy showed that the ulcer had healed. T h e stricture was successfully dilated to 43 F with E d e r - P u e s t o w dilators without complications. H e was discharged on phosphonoformate 30 m g / k g x2 hourly. His Hickman line was removed following an episode of line sepsis, and phosphonoformate was stopped after 3 months of maintenance treatment. H e remained free ofoesophageal symptoms, but died of Staphylococcus lugdunensis endocarditis 8 months after the initial diagnosis. Oesophageal ulceration due to C M V is an u n c o m m o n problem in patients with H I V infection. Only four cases were identified among 750 seropositive patients in a year in one study. 1 T r e a t m e n t with ganciclovir or phosphonoformate produces symptomatic relief in up to 80 % patients, 2 but the prognosis is poor, with a median survival of 4-6 months. 2.3 T h e r e is no consensus about the need for maintenance treatment after induction therapy. Some patients have achieved relatively long remissions after initial treatmentfl but relapse of ulceration has also been reportedfl Many patients die before their oesophageal disease relapses, and a policy of starting all patients on maintenance therapy would involve treating many of these unnecessarily. Our patient was started on maintenance treatment because he had particularly prolonged and severe symptoms, and the benefits were felt to outweigh the disadvantages. Nevertheless complications of this treatment contributed to his death. Initial treatment courses in C M V oesophagitis may need to be prolonged, and there is a need for formal assessment of the role of maintenance treatment. Stricture formation due to C M V has been reported in one patient previously, ~ but we believe this to be the first report of successful dilatation of such a stricture. Stricture dilatation appears to be safe and is worth considering despite the poor overall prognosis in advanced A I D S as it produced a marked symptomatic improvement in our patient.
Department of Medicine, Middlesex Hospital, London W I N 8AA, U.K.
Duncan Churchill Jesse Kenton-Smith Adam Malin
Letters to the Editor
I09
References I. Jacobson MA, O'Donnell JJ, Porteous D, Brodie HR, Feigal D, Mills J. Retinal and gastrointestinal disease due to cytomegalovirus in patients with the acquired immune deficiency syndrome : prevalence, natural history, and response to ganciclovir therapy. Q ff Med I988; 67: 473-486. 2. Wilcox CM, Diehl DL, Cello JP, Margaretten W, Jacobson MA. Cytomegalovirus esophagitis in patients with AIDS. Ann Intern Med I99o; II3:589-593 3. Connolly GM, Hawkins, D, Harcourt-Webster JN, Parsons PA, Husain OAN, Gazzard BG. Oesophageal symptoms, their causes, treatment, and prognosis in patients with the acquired immunodeficiency syndrome. Gut I989; 3o: Io33-ro39. Kikuchi's disease associated with E p s t e i n - B a r r virus infection Accepted for publication 2 January I992 Sir, Kikuchi's disease (KD) is a cause of benign lymphadenitis known to have a worldwide distribution, reports of nearly 450 cases having been published. 1 We did not know, however, of any case in Portugal until a 15-year-old caucasian boy presented to us with pyrexia of unknown origin (PUO) and liver dysfunction. T h e patient gave a 4 weeks' history of daily fever (39-40 °C), enlarged cervical lymph nodes, night sweats, chills, malaise, anorexia and a 4 kg loss of weight. H e had complained of a sore throat during the first week and had developed a skin rash after a I-day course of amoxicillin. H e denied any other symptoms. His epidemiological history was unremarkable. On examination he had an axillary temperature of 39 °C and a single 3 c r u x 4 cm, non-tender cervical lymph node and a few small supraclavicular nodes on the left side. His physical examination was otherwise unremarkable. Laboratory investigation showed leucopenia (3"45 x IO9/1) with relative lymphocytosis (I'7 x Io9/1), E S R of 86 m m / h (Westergren), serum lactate dehydrogenase ( L D H ) IZ87 IU/1 (n = I60-33o IU/1), serum alanine aminotransferase of I78 IU/1 (n = 0--29 IU/1) and aspartate aminotransferase of i2I IU/1 (n = 0-25 IU/1). All blood, urine stool and bone marrow bacterial cultures were negative. Acid-fast bacilli were not found in sputum, bone marrow, urine and gastric washings. His Mantoux test was negative. Bone marrow aspirate and biopsy, autoimmune studies, chest roentgenography and abdominal ultrasonography were unremarkable. Rose Bengal and Paul-Bunnell tests were negative. No yersinia antibodies were detected and toxoplasma and cytomegalovirus I g M and I g G titres were not raised. However, I g M VCA antibody to Epstein-Barr virus (EBV) was detected. During the first week in the ward the patient maintained a fever (39-40 °C), with progressive weight loss (5 kg), increased serum aminotransferases and L D H , sustained leucopenia and relative lymphocytosis. In the second week, a biopsy of a supraclavicular lymph node showed extensive necrosis within the cortical and paracortical areas, with karyorrhectic debris and no neutrophil polymorphonuclear leucocytes. No tubercle bacilli were seen. T h e specimen was reported to show benign necrotising lymphadenitis, consistent with Kikuchi's disease. T r e a t m e n t with naproxen (25o mg b.d.), resulted in p r o m p t lysis of the fever. F o u r weeks later, when he was discharged, he was apyrexial, regaining weight and had resolving cervical adenopathy. His serum aminotransferase values were improving. At a 6 m o n t h follow-up he was well, had no adenopathy and his aminotransferases were normal. Kikuchi's disease is probably misdiagnosed in our country, and therefore, the clinical manifestations of this benign lymphadenitis should be more widely known and lymph node biopsy more readily performed. Kikuchi's disease with abnormal liver
Io8
Oesophageal stricture complicating cytomegalovirus ulceration in a patient with AIDS
Accepted for publication 27 November I99I Sir, Cytomegalovirus (CMV) infection is a well-recognised cause of painful oesophageal ulceration in patients with H I V infection. We report a case in which an oesophageal ulcer was complicated by the development of a stricture which was successfully treated. A z3-year-old HIV-positive homosexual man presented with retrosternal pain on swallowing. H e had been treated for Pneumocytis carinii pneumonia 2 years previously. T w o months before admission he had developed oral candidiasis, which responded to oral fluconazole. Despite an increased dose offluconazole he continued to experience pain. Endoscopy demonstrated a typical large mid-oesophageal ulcer, and biopsies showed numerous C M V inclusions. T r e a t m e n t was started with Iv ganciclovir (io m g / k g / 2 4 h via a Hickman line, and parenteral diamorphine was required for pain control. After 25 days there had been some improvement in his symptoms, but neutropenia then developed and ganciclovir was stopped. Another endoscopy showed a partly healed ulcer, with stricturing of the mid-oesophagus, and upper oesophageal candidiasis. T h e dose of fiuconazole was again increased, and he was commenced on phosphonoformate 6o m g / k g 8 hourly for 3 weeks. Following this his pain disappeared, but he developed dysphagia for solid food. A further endoscopy showed that the ulcer had healed. T h e stricture was successfully dilated to 43 F with E d e r - P u e s t o w dilators without complications. H e was discharged on phosphonoformate 30 m g / k g x2 hourly. His Hickman line was removed following an episode of line sepsis, and phosphonoformate was stopped after 3 months of maintenance treatment. H e remained free ofoesophageal symptoms, but died of Staphylococcus lugdunensis endocarditis 8 months after the initial diagnosis. Oesophageal ulceration due to C M V is an u n c o m m o n problem in patients with H I V infection. Only four cases were identified among 750 seropositive patients in a year in one study. 1 T r e a t m e n t with ganciclovir or phosphonoformate produces symptomatic relief in up to 80 % patients, 2 but the prognosis is poor, with a median survival of 4-6 months. 2.3 T h e r e is no consensus about the need for maintenance treatment after induction therapy. Some patients have achieved relatively long remissions after initial treatmentfl but relapse of ulceration has also been reportedfl Many patients die before their oesophageal disease relapses, and a policy of starting all patients on maintenance therapy would involve treating many of these unnecessarily. Our patient was started on maintenance treatment because he had particularly prolonged and severe symptoms, and the benefits were felt to outweigh the disadvantages. Nevertheless complications of this treatment contributed to his death. Initial treatment courses in C M V oesophagitis may need to be prolonged, and there is a need for formal assessment of the role of maintenance treatment. Stricture formation due to C M V has been reported in one patient previously, ~ but we believe this to be the first report of successful dilatation of such a stricture. Stricture dilatation appears to be safe and is worth considering despite the poor overall prognosis in advanced A I D S as it produced a marked symptomatic improvement in our patient.
Department of Medicine, Middlesex Hospital, London W I N 8AA, U.K.
Duncan Churchill Jesse Kenton-Smith Adam Malin
Letters to the Editor
I09
References I. Jacobson MA, O'Donnell JJ, Porteous D, Brodie HR, Feigal D, Mills J. Retinal and gastrointestinal disease due to cytomegalovirus in patients with the acquired immune deficiency syndrome : prevalence, natural history, and response to ganciclovir therapy. Q ff Med I988; 67: 473-486. 2. Wilcox CM, Diehl DL, Cello JP, Margaretten W, Jacobson MA. Cytomegalovirus esophagitis in patients with AIDS. Ann Intern Med I99o; II3:589-593 3. Connolly GM, Hawkins, D, Harcourt-Webster JN, Parsons PA, Husain OAN, Gazzard BG. Oesophageal symptoms, their causes, treatment, and prognosis in patients with the acquired immunodeficiency syndrome. Gut I989; 3o: Io33-ro39. Kikuchi's disease associated with E p s t e i n - B a r r virus infection Accepted for publication 2 January I992 Sir, Kikuchi's disease (KD) is a cause of benign lymphadenitis known to have a worldwide distribution, reports of nearly 450 cases having been published. 1 We did not know, however, of any case in Portugal until a 15-year-old caucasian boy presented to us with pyrexia of unknown origin (PUO) and liver dysfunction. T h e patient gave a 4 weeks' history of daily fever (39-40 °C), enlarged cervical lymph nodes, night sweats, chills, malaise, anorexia and a 4 kg loss of weight. H e had complained of a sore throat during the first week and had developed a skin rash after a I-day course of amoxicillin. H e denied any other symptoms. His epidemiological history was unremarkable. On examination he had an axillary temperature of 39 °C and a single 3 c r u x 4 cm, non-tender cervical lymph node and a few small supraclavicular nodes on the left side. His physical examination was otherwise unremarkable. Laboratory investigation showed leucopenia (3"45 x IO9/1) with relative lymphocytosis (I'7 x Io9/1), E S R of 86 m m / h (Westergren), serum lactate dehydrogenase ( L D H ) IZ87 IU/1 (n = I60-33o IU/1), serum alanine aminotransferase of I78 IU/1 (n = 0--29 IU/1) and aspartate aminotransferase of i2I IU/1 (n = 0-25 IU/1). All blood, urine stool and bone marrow bacterial cultures were negative. Acid-fast bacilli were not found in sputum, bone marrow, urine and gastric washings. His Mantoux test was negative. Bone marrow aspirate and biopsy, autoimmune studies, chest roentgenography and abdominal ultrasonography were unremarkable. Rose Bengal and Paul-Bunnell tests were negative. No yersinia antibodies were detected and toxoplasma and cytomegalovirus I g M and I g G titres were not raised. However, I g M VCA antibody to Epstein-Barr virus (EBV) was detected. During the first week in the ward the patient maintained a fever (39-40 °C), with progressive weight loss (5 kg), increased serum aminotransferases and L D H , sustained leucopenia and relative lymphocytosis. In the second week, a biopsy of a supraclavicular lymph node showed extensive necrosis within the cortical and paracortical areas, with karyorrhectic debris and no neutrophil polymorphonuclear leucocytes. No tubercle bacilli were seen. T h e specimen was reported to show benign necrotising lymphadenitis, consistent with Kikuchi's disease. T r e a t m e n t with naproxen (25o mg b.d.), resulted in p r o m p t lysis of the fever. F o u r weeks later, when he was discharged, he was apyrexial, regaining weight and had resolving cervical adenopathy. His serum aminotransferase values were improving. At a 6 m o n t h follow-up he was well, had no adenopathy and his aminotransferases were normal. Kikuchi's disease is probably misdiagnosed in our country, and therefore, the clinical manifestations of this benign lymphadenitis should be more widely known and lymph node biopsy more readily performed. Kikuchi's disease with abnormal liver
