Histopathology of Human Immunodeficiency Virus-Associated Esophageal Disease

We evaluated the histopathologic features of the esophageal mucosa in 88 patients seropositive for human immunodeficiency virus (HIV). All patients had an upper endoscopy because of esophageal symptoms. Forceps biopsies and brushings of the esophagus were examined histologically and cytologically for evidence of viral, fungal, and mycobacterial infections: in addition, biopsies and brushings were cultured for cytomegalovirus and herpes simplex. Esophageal inflammation (acute or chronic) was graded 0 through 3. Twenty-one patients (24%) had a normal endoscopy; none displayed high grade (grade 2 and 3) acute inflammation and only two (9.5%) had high grade chronic inflammation in the esophagus. Moreover, no fungi or viral inclusions were seen in samples from these patients. Eleven patients (12%) had an abnormal esophageal mucosa but no pathogen detected and were categorized as "idiopathic esophagitis." The percent with high-grade inflammation (27%) was not significantly different from the normal group. Fifty-six patients (64%) had an infectious diagnosis. Forty-six percent had Candida , 16% had viral esophagitis alone, and one patient had Kaposi's sarcoma. Infections were associated with high-grade acute and chronic inflammation in 53% and 47% of patients, respectively. The location of the infiltrate did not predict the type of infection. In conclusion, if esophagoscopy is normal in patients with HIV infection and esophageal symptoms, a biopsy is not necessary.     

Comparison of Two Corticosteroid Regimens for the Treatment of HIV-associated Idiopathic Esophageal Ulcer

Objective: Although corticosteroids appear to be efficacious in the treatment of HIV-associated idiopathic esophageal ulcer, the optimal regimen remains undeflned. Methods: Over a 41-month period, all patients with idiopathic esophageal ulcer defined hy clinical, endoscopic, and pathological criteria were treated with a defined corticosteroid regimen. The initial 12 patients were treated with prednisone, 40 mg daily tapering to 10 mg/wk, for a 4-wk course of therapy, and 24 suhsequent patients were treated with 40 mg daily for 2 wk. All patients were followed clinically, with most patients undergoing endoscopy at the completion of therapy and for recurrent esophageal symptoms. Results: A symptomatic response was seen in 11 patients (92%) in the 4-wk regimen and 23 patients (96%) in the 2-wk regimen. A complete symptomatic response was seen in 10 patients (83%) in the 4-wk regimen, compared with 18 patients (75%) in the 2-wk regimen. A partial response was seen in the 2-wk regimen more often (five patients) than in the 4-wk regimen. Long-term follow-up in responders revealed relapse in two of nine patients (22%) in the 4-wk regimen compared with 12 of 23 (52%) in the 2-wk regimen ( p = 0.10). Relapse occurred within 9 wk in all hut one patient and was heralded hy recurrent odynophagia. The prednisone regimen was well-tolerated, although opportunistic infections occurred in seven patients during or within 1 month of therapy. Conclusions: Prednisone therapy is highly efficacious for the treatment of idiopathic esophageal ulcer resulting in a rapid clinical response and endoscopic healing. A 2-wk course of therapy may he associated with a higher relapse rate than a 4-wk regimen. Relapse frequently occurs, usually within 2 months of therapy. The side effects proflle appears acceptable.     

Actinomyces Infection of a Cytomegalovirus Esophageal Ulcer in Two Patients with Acquired Immunodeficiency Syndrome

Esophageal disease is a significant cause of morbidity among patients with the acquired immunodeflciency syndrome (AIDS). Many organisms have been implicated in the pathogenesis of dysphagia and odynophagia. We describe a unique presentation of actinomyces esophageal infection in two homosexual male patients with AIDS and biopsy proven CMV esophagitis. After failure of esophagitis to resolve with ganciclovir or foscarnet therapy, the patients underwent repeat endoscopy and were subsequently found to have a secondary infection of the ulcers by Actinomyces . Treatment with intravenous penicillin G resulted in symptomatic and histopathological resolution of esophageal disease. This appears to be the first report of Actinomyces infection of esophageal ulcers in AIDS patients, possihiy a commonly overlooked diagnosis.     

Idiopathic Anorectal Ulceration in Patients with Human Immunodeficiency Virus Infection

Opportunistic infections and neoplasms of the anorectum have heen reported in patients with human immunodeficiency virus (HIV) infection. More recently, idiopathic alcerative lesions of the colon and rectum have been described. At our center over a 3-yr period, four patients were identified with ulcerative lesions of the rectum and/or anus that remained idiopathic despite an extensive clinical, serologic, and histopathologic evaluation. Three patients had the acquired immunodeficiency syndrome, and in one anorectal disease was the index manifestation of HIV infection. Only one of the patients had recently engaged in receptive anal intercourse. The presenting complaints were gastrointestinal hieeding in two, which was severe in one, and/or anorectal pain. Multiple colonoscopic evaluations with hiopsy of the distal colorectum documented a solitary ulcer of the rectum in one, solitary ulcer involving the anorectum in two, and multiple ulcers of the rectum and anorectnm in one. In three patients, colonoscopy to the cecum demonstrated no additional lesions. In patients with HIV infection, ulcerative lesions of the anorectum may remain unexplained despite an exhaustive evaluation. The etiology of these lesions, as well as appropriate therapy, remains to be determined.     

Idiopathic Esophageal Varices

A 54-yr-old man with idiopathic esophageal varices is presented. Despite extensive work-up, no etiology for the varices was found. Previously published cases of idiopathic esophageal varices are reviewed.     

Co-Infection with Paracoccidioidomycosis and Human Immunodeficiency Virus: Report of a Case with Esophageal Involvement

Paracoccidioiodomycosis (PCM) is a systemic and deep mycosis endemic in Latin America, especially in Brazil. In patients infected with human immunodeficiency virus (HIV), PCM can manifest with prominent involvement of the reticuloendothelial system. There are no reports in the literature of esophageal involvement by PCM in that population. We report a case of PCM with pulmonary and esophageal involvement without radiologic evidence of an esophageal-bronchial fistula in an HIV-infected patient.Paracoccidioidomycosis (PCM) is a systemic mycosis endemic in Latin American countries, with higher prevalence in Brazil, especially in the southeastern, southern, and central-western regions.^1,2 Patients are infected by inhaling the conidia of Paracoccidioides brasiliensis, a thermodimorphic fungus, with the occurrence of a primary pulmonary infection.^3,4 Infection is usually controlled by the cell-mediated immune response and is asymptomatic. However, viable forms may persist inside the healed primary lesion.⁵In persons infected with human immunodeficiency virus (HIV), PCM occurs primarily in the juvenile form with prominent involvement of the reticuloendothelial system, although pulmonary and oral mucosa involvement, more typical of the chronic form, frequently coexist. The disease occurs mainly in patients with T CD4⁺ lymphocyte counts less than 200 cells/mm³.^3,6,7There are no reports in the literature of esophageal involvement for PCM in HIV-infected patients. We report a case of pulmonary and esophageal PCM affecting an HIV-infected patient with no evidence of an esophageal-bronchial fistula or of associated skin or oral lesions and/or lymphadenopathy.A 55-year-old, heterosexual, white man who was born and lived in the southeastern region of the state of São Paulo, Brazil, had a serologic diagnosis of HIV infection 10 years earlier. The patient reported postprandial epigastric and retrosternal pain and odynophagia of 20 days duration, daily evening fever of one-month duration, dry cough, and a weight loss of 15 kg over three 3 months. He also reported a history of treatment for pulmonary tuberculosis identified at the time of serologic diagnosis of HIV by culture of three sputum samples from which Mycobacterium tuberculosis was isolated.Physical examination showed a patient in good general condition, without adenomegalies, who had hepatosplenomegaly and an inguinal hernia on the right side. A chest radiograph showed diffuse reticulonodular opacification in the upper thirds of both lungs. A rounded calcified nodule 0.8 cm in diameter was detected in the lower third of the left lung. He also showed obliteration of the left costophrenic sinus caused by a pleural reaction.After admission to the hospital, the presence of acid-fast bacilli was tested in sputum; results were negative in three samples. The T CD4⁺ lymphocyte count was 269 cells/µL and serum HIV viral load was 219,829 copies/mL (log 5.3404). Upper digestive endoscopy showed a deep ulcer, without fibrin, in the distal third of the esophagus (Figure 1). Histologic examination of a biopsy specimen of this esophageal lesion by staining with hematoxylin and eosin showed an area of ulceration covered with a fibrin-leukocyte exudate. Mixed inflammatory infiltrate consisting of numerous neutrophils, lymphocytes, and macrophages was observed at the level of the lamina propria. There was no evidence of granulomas, but many small, rounded structures with a birefringent halo were detected throughout connective tissue (Figure 2). Staining with Gomori methenamine and silver identified the morphology of these structures and showed numerous rounded, double-walled fungal cells of various sizes, with single or multiple budding and focal rudder-shaped structures morphologically consistent with Paracoccidioides brasiliensis (Figure 3). Results of a test for acid-fast bacilli by using the Ziehl-Neelsen method were negative. Serum counterimmunoelectrophoresis was reactive for PCM, with titers ranging from 1:16 to 1:32, but not reactive for histoplasmosis, cryptococcosis or aspergillosis.Open in a separate windowFigure 1.Endoscopic view of a deep ulcer in the distal third of the esophagus of the patient. This figure appears in color at www.ajtmh.org.Open in a separate windowFigure 2.Esophageal biopsy of the patient, showing rounded fungal structure with double-refringent walls (arrows) (hematoxylin and eosin stained, magnification ×1,000) and detail of fungal structure. This figure appears in color at www.ajtmh.org.Open in a separate windowFigure 3.Esophageal biopsy of the patient, showing numerous rounded fungal structures with single and multiple budding (arrows) (Gomori methenamine and silver staining, magnification ×400) and detail of multiple budding. This figure appears in color at www.ajtmh.org.The patient received specific treatment for PCM, and showed remission of signs and symptoms and healing of the esophageal ulcer. Ten months later, he returned for surgical correction of the right inguinal hernia. The patient then had suture dehiscence, Fournier syndrome, bilateral pneumonia, septic shock, and died. An autopsy showed an esophagus, stomach, and small and large intestines without abnormalities. The lungs had a confluent bronchopneumonic process with a dense intra-alveolar neutrophil exudate, rare rudiments of granulomas, and occasional multinucleated giant cells. Staining with Gomori methenamine and silver showed numerous rounded fungi of various sizes amid areas of necrosis, with single or predominantly multiple budding, compatible with pulmonary PCM.The estimated prevalence of PCM among HIV-infected persons seen at the Division of Infectious and Tropical Diseases of the University Hospital, Faculty of Medicine of Ribeirão Preto, University of São Paulo, is 1.4%, which is comparable to the prevalence of 1.5% reported for the state of Mato Grosso do Sul in the central-western region of Brazil.⁵ This prevalence is relatively low among patients with acquired immunodeficiency syndrome (AIDS) in Brazil and in other countries in South America when compared with other mycoses such as histoplasmosis and cryptococcosis. This finding may be caused by other factors such as epidemiologic differences between HIV infections, which is a phenomenon in large urban centers (although the AIDS epidemic has spread to smaller cities and rural areas in Brazil), and infection by P. brasiliensis, which affects mainly small agricultural communities.^4,5 Another factor is the prophylactic use of trimethoprim-sulfamethoxazole for infection with Pneumocystis jirovecii, which is also effective against P. brasiliensis.⁴ In experimental studies, Chiarella and others⁸ demonstrated an important role of CD8⁺ T cells in immunoprotection against pulmonary PCM in mice. Because patients infected with HIV have a relative increase in CD 8⁺ T cells and a decrease in CD4⁺ T cells, this factor may explain the low prevalence of PCM among patients with AIDS.Our patient was from a town in the state of São Paulo that had mainly rural activity. Thus, PCM may have been the result of reactivation of a latent fungal infection acquired in the past and triggered by acquisition of HIV infection.⁹Esophageal involvement by PCM is extremely rare. Reports by Oliveira and others¹⁰ and Ziliotto and others¹¹ of esophageal P. brasiliensis infection refer to patients not infected with HIV. Endoscopic findings reported by these investigators were stenosing and vegetating lesions suggestive of neoplasia located in the upper and middle third of the esophagus. In contrast to reports of esophageal PCM affecting immunocompetent persons, our patient had an ulcerated lesion in the distal third of the esophagus. Esophageal ulcers are important causes of morbidity among patients with AIDS. The agents most frequently observed in these lesions are cytomegalovirus, Candida sp., and herpes simplex virus. Infections with other fungi such as Cryptococcus neoformans and Histoplasma capsulatum are less common.¹² There are no reports in the literature of esophageal involvement by P. brasiliensis in HIV-infected persons.The etiologic diagnosis of an esophageal ulcer in our patient was based on the combined results of complementary methods. Histologic examination of biopsy specimens from the esophageal ulcer after staining with hematoxylin and eosin showed rounded structures throughout connective tissue in the lamina propria. It was possible to obtain a good definition of the typical characteristics of PCM only by staining with Gomori methenamine and silver, which showed many rounded fungi of various sizes with single and multiple budding and focal rudder-shaped structures.The differential diagnosis included Histoplasma capsulatum var capsulatum, Blastomyces dermatitidis, Cryptococcus neoformans, and Coccidioides immitis.¹³ The multiple budding characteristic of PCM was important for the histopathologic diagnosis reached in agreement with these features. Serum counterimmunoelectrophoresis was reactive for PCM although at low titers (1:16–1:32), a feature similar to that reported by Paniago and others and probably caused by B cell dysfunction observed in HIV-infected patients.⁴ A chest radiograph showed a diffuse reticulonodular infiltrate, a change usually observed in pulmonary lesions caused by PCM, and a calcified nodular image of a residual aspect in the left lung (probably caused by previously treated pulmonary tuberculosis).The autopsy findings showed, in addition to injuries caused by septic shock, pulmonary involvement caused by an abscessed bronchopneumonic process, which when examined histologically by staining with Gomori methenamine and silver, showed many fungal forms with morphologic characteristics identical to those observed in the esophageal biopsy specimens, and rare rudimentary granulomas intermingled with areas of necrosis, and parenchymatous hemorrhage. In our patient, the cause of death could be attributed to a combination of factors mainly caused by specific immunodeficiency associated with HIV infection, which favored the persistence of pulmonary PCM, and by the septic condition of the patient. These aspects coincide with those reported by Paniago and others, who demonstrated a high mortality caused by PCM in HIV-infected persons.⁴Mortality data from the State Foundation of Analysis of Data of São Paulo, Brazil, during 1985–2005 showed 1,950 deaths; PCM was the cause of death in 59.7% of the cases and was an associated cause of death in 40.3%. Malignant neoplasias and AIDS were the main causes of death in which PCM was reported as an associated cause.¹⁴ Thus, similar to Morejón and others,⁵ we also believe that PCM should be classified as an opportunistic disease and included among the diseases that define AIDS in HIV-infected persons living in Latin America,⁵ although new studies are needed to corroborate these data. We conclude that PCM is a disease of great relevance in patients with HIV/AIDS in Latin America, especially in Brazil. It also shows peculiar clinical characteristics that should be considered in the possible etiology of esophageal ulcers in HIV-infected persons.     

Prospective Evaluation of Oropharyngeal Findings in Human Immunodeficiency Virus-infected Patients with Esophageal Ulceration

Objective: Although the presence of oropharyngeal (OP) candidiasis plays an important role in the evaluation of human immunodeficiency virus (HIV)-infected patients with esophageal symptoms, there is little information on the utility of OP findings in patients with esophageal ulceration.
Methods: Over a 54-month period, all HIV-infected patients with esophageal ulceration had careful inspection of the oropharynx at the time of presentation with esophageal complaints and at endoscopy. HIV-infected patients without esophageal ulceration undergoing endoscopy during the last 30 months had OP findings similarly documented. OP ulceration was not routinely biopsied. The cause of esophageal ulceration was determined based on clinical, en-doscopic, and histopathological findings.
Results: Of the 124 patients identified with esophageal ulcer, 14 (11%) had coexistent OP ulceration: herpes simplex virus, four; idiopathic esophageal ulcer, four; cytomegalovirus, three; herpes simplex virus and cytomegalovirus, two; idiopathic and cytomegalovirus, one. Four patients had OP ulcer without esophageal ulcer; only one of these patients had esophageal symptoms. All OP lesions healed with therapy for the esophageal ulcer. Twenty-eight patients with esophageal ulcer had OP candidiasis (23%); 21 of these patients (75%) also had Candida esophagitis. The sensitivity and specificity of OP ulcer for esophageal ulcer were 11% and 97%, respectively. The positive and negative predictive values of OP candidiasis for esophageal candidiasis were 90% and 82%, respectively.
Conclusions: OP ulceration is uncommon in patients with esophageal ulceration, with the exception of herpes simplex virus esophagitis. OP candidiasis is common in patients with underlying esophageal ulcer, potentially resulting in diagnostic confusion. OP candidiasis appears to be a moderately useful diagnostic marker for Candida esophagitis.     

Coinfection: Helicobacter pylori/Human Immunodeficiency Virus

To compare H. pylori infection prevalence and gastric mucosa damage in HIV-infected and non-HIV-infected patients, gastric biopsies were systematically taken in 209 individuals who underwent upper Gl endoscopy (102 HIV-infected and 107 non-HIV-infected). H. pylori was found in 42 (41.1%) HIV-infected patients and in 53 (49.5%) non-HIV patients (P = 0.22, chi2 = 1.47, NS). In HIV-positive patients infected with H. pylori the mean CD4 count was higher than in HIV-positive patients without H. pylori (364 and 228 cells/mm3, respectively; P = 0.0001). H. pylori gastritis was more severe in the HIV-positive group (chi2 = 15.02, P = 0.0001). The frequency of H. pylori in gastric mucosa in HIV-infected and non-HIV patients was similar. HIV-infected patients with H. pylori had a higher mean CD4 count than HIV-infected individuals without H. pylori. Gastric lesions associated with H. pylori were more severe in the HIV-positive population.     

A Case of Kissing Esophageal Ulcers of Unknown Etiology

Abstract: A 41-year-old man with esophageal ulcers of unknown etiology is reported. Endoscopic examination was performed for evaluation of swallowing difficulty. The lesions were located in the middle portion of the esophagus. There were no findings correlating with either peptic ulcer or reflux esophagitis. In addition, he had no history of provoking factors, such as bacterial, chemical, traumatic or physical agents. Although the cause of these lesions was not clarified, we diagnosed acute benign ulcers following histological examination of the biopsy specimens. The patient was treated with a proton pump inhibitor and sodium alginate. The dysphagia and other symptoms subsided promptly, and the lesions healed completely within three weeks. We discuss herein the relationship between endoscopic findings and the etiology of kissing ulcer.     

Return of Esophageal Peristalsis after Nifedipine Therapy in Patients with Idiopathic Esophageal Achalasia

This study was carried out to demonstrate the possible return of esophageal peristalsis in patients affected by esophageal achalasia chronically treated with sublingual nifedipine and to investigate which parameters are correlated with the return of peristalsis. Thirty-two patients were treated with sublingual nifedipine 10-20 mg taken 30 min before meals. A clinical and manometric evaluation was performed before and after 6 months of therapy. Before treatment, in no patient was peristaltic activity recorded. After 6 months, peristalsis was observed in six patients. In this group, no pretreatment manometric parameter was different from that of the remaining achalasic patients; only the clinical history of dysphagia was significantly shorter (p < 0.001) and the esophageal diameter significantly less (p < 0.001). In conclusion, chronic treatment with sublingual nifedipine can induce a return of esophageal peristalsis in patients with a short clinical history of disease and slightly dilated esophagus.     

Challenges Associated with Management of Buruli Ulcer/Human Immunodeficiency Virus Coinfection in a Treatment Center in Ghana: A Case Series Study

The synergy between Mycobacterium tuberculosis infection and human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome is well established but not so in Buruli ulcer (BU). We screened confirmed BU cases for HIV infection and followed seven BU/HIV-coinfected patients. Management of BU/HIV was based on the World Health Organization guidelines and patient condition. The HIV positivity among BU patients (8.2%; 11/134) was higher compared with that of general patients attending the facility (4.8%; 718/14,863; P = 0.07) and that of pregnant women alone (2.5%; 279/11,125; P = 0.001). All seven BU/HIV-coinfected cases enrolled in the study presented with very large (category III) lesions with four having multiple lesions compared with 54.5% of category III lesions among HIV-negative BU patients. During the recommended BU treatment with streptomycin and rifampicin (SR) all patients developed immune infiltrates including CD4 T cells in their lesions. However, one patient who received antiretroviral therapy (ART) 1 week after beginning SR treatment developed four additional lesions during antibiotic treatment, while two out of the four who did not receive ART died. Further evidence is required to ascertain the most appropriate time to commence ART in relation to SR treatment to minimize paradoxical reactions.     

Idiopathic orthostatic tachycardia. Etiology, diagnosis and treatment

For nearly a century, physicians have been aware of a syndrome consisting of a relatively stereotyped presentation, usually in young patients, who complain of fatigue, malaise and effort intolerance, sometimes of trembling and weakness of the lower limbs. This is associated with an excessive tachycardia in the orthostatic position. This syndrome has recently been called idiopathic orthostatic tachycardia. The tilt test has enabled "quantification" of normal responses. Patients complaining of the symptoms described above and which, during the first minutes of orthostatism, increase their heart rates by more than 30 beats per minute or attain a rate of at least 110/min, are considered to be suffering from this syndrome. The physiopathology is not clear but, globally, there seems to be two sub-groups, the first considered to be a partial dysautonomic disorder and the second, the result of hypersensitivity of the beta-receptors. Besides the tilt test, the diagnosis can also be presumed after an excessive tachycardia response to an intravenous infusion of 1 microgram/min of isoprenaline. The treatment of these patients is uncertain as there is no single approach which is always effective. In addition to "simple" but essential advice, a number of drugs may be used although there is no means of predicting the efficacy of the result in a given patient. A major principle should be emphasised: ablation of the sinus node for inappropriate tachycardia may eliminate the only compensatory mechanism of autonomic dystonia and make the patients even more symptomatic than they were.     

Manometric Characteristics in Idiopathic and Reflux-Associated Esophageal Spasm

Ancillary manometric findings, e.g., high amplitude contractions, repetitiveness, or elevated lower esophageal sphincter (LES) pressure, have been reported in diffuse esophageal spasm (DES). However, two recent changes in DES have been noted: 1) it has been redefined as increased simultaneous contractions, with intermittent peristalsis, and 2) there has been more attention to reflux-associated DES. Therefore, our aims were to characterize the ancillary findings in currently defined DES and to determine whether these occurred in both idiopathic and reflux-associated DES. Records of 31 patients with DES (greater than 25% simultaneous contractions) were reviewed. Independent of manometry, some patients could be subclassified as idiopathic (N = 7; no heartburn; normal endoscopy or acid perfusion test) or reflux-associated (N = 10; heartburn; positive endoscopy). Both low and high LES pressures and contraction amplitudes were seen. Repetitive contractions were seen in nearly all patients, and segmental aperistalsis, dropped waves, or distally nonpropagated waves were seen in more than half. These findings were generally observed in both types of DES. This study of DES 1) confirms the high prevalence of repetitive contractions, 2) deemphasizes high LES pressure and contraction amplitude, 3) extends the findings to include other types of peristaltic dysfunction, and 4) indicates that manometric findings per se do not allow clear differentiation of idiopathic from reflux-associated DES.