- Email: [email protected]
Pleural Tuberculosis in HIV-Infected Patients
illary blood depending on the distribution of ventilation and blood flow. The systemic arterial blood oxygen tension is a result of a mixture of all end-capillary blood derived from the lung combined with blood from anatomic shunts. Therefore, there are many true alveolar-arterial oxygen "gradients" that cannot be represented accurately by one number. The authors are not describing a true physical gradient but rather, an abstract mathematical difference. I would use the term alveolar-arterial oxygen "difference" for the mathematical subtraction of the measured arterial oxygen tension from the idealized alveolar oxygen tension.
Adults and the Elderly," an article published in Chest in July 1994. They summarized that the young "were more likely to have hemoptysis, fever, and cough ... . " Although the reference probably did not appear in their literature search, there has been a previous description of this phenomenon. The consumption of young men that are in the flower of their age, when the heat of blood is yet brisk, and therefore more disposed to a feverish fermentation, is for the most part acute. But, in old men, where the natural heat is decayed, it is more chronicaJ.l
It is reassuring that some clinical aphorisms have enduring value.
Carl M. Kirsch, MD, FCCP, Division of Respiratory/ Critical Care Medicine , Santa Clara Valley Medical Center, San jose, California
Fistulas Do Not Always Cause Pericardial Effusion
Michael D. Iseman, MD, FCCP, National jewish Center for Immunology and Respiratory Medicine, Denver, Colorado REFERENCE
1 Morton R. Phthisiologia. 1689
To the Editor: Mahaisa variya and colleagues report (Chest 1994; 106:1285-88) on an atrial-esophageal fistula, which adds another interesting case to the roughly 65 comparable cases reported, usually due to benign esophageal disease and sometimes involving the pericardium. The authors note: "The adherence of the esophagus and the myocardium, therefore, provides a potential for fistula development." This needs correcting: the esophagus does not adhere to the myocardium ; the pericardium is interposed between the esophagus and the adjacent cardiovascular structures from the level of T-3 to T-11 1 The probable reason that a fistula does not always cause pericardia! effusion, because it must perforate the interposed pericardium , is the unusual structure of the pericardium over the left atrium-a basketwork of fibers in the parietal pericardium, which tightly clasp the left atrium 2 This is also the reason why none but very large pericardia! effusions under pressure penetrate behind that structure, and why patients stabbed in the back through the left atrium tend to bleed into the pleural cavity rather than the pericardium. These remarks are not made to criticize but rather to amplify a nice report.
David H. Spodick, MD, DSc, Cardiology Division , Saint Vincent Hospital, Worcester, Massachusetts REFERENCES
Spodick DH. Macro- and Micro-physiology and anatomy of the pericardium. Am Heart J 1992; 124:1046-51 2 Spodick DH. The normal and diseased pericardium: current concepts of pericardia! physiology, diagnosis and treatment. J Am Coli Cardiol1983; 1:240-51
Tuberculosis in Young Adults and the Elderly To the Editor: Korzeniewska-Kosela and colleagues (Chest 1994; 106:28-32) reported their observations contrasting "Tuberculosis in Young
Pleural Tuberculosis in HIV-Infected Patients To the Editor: We have read with great interest the article by Relkin et al (Chest 1994; 105:1338-41), which appeared in Chest in May 1994, about the characteristics of pleural tuberculosis in patients infected by HIV. In this report the authors found that patients with pleural tuberculosis and HIV infection, had compared with a group of HIV -negative patients, significantly fewer positive tuberculin skin tests, more acid-fast bacteria identifiable in pleural tissue, higher proportion of positive Mycobacterium tuberculosis (MT) cultures of sputum and pleural biopsy (PB), and similar pleural biopsy histologic conditions. It has been suggested that delayed hypersensitivity plays a major role in the pathogenesis of tuberculous pleural effusion, 1 and T-lymphocytes play a central role in cell-mediated defences against MT. 2 Because HIV infection results in specific depletion of CD4 cells, we might therefore expect to find more acid-fast bacteria identifiable in pleural tissue, sputum, and pleural fluid (PF), and fewer granulomas on PB specimens. To evaluate this hypothesis, we retrospectively studied all cases with a definitive diagnosis of pleural tuberculosis from 1986 through December 1991 at our hospital. Inclusion criteria in the study were (1) positive culture of PF, PB specimen(s), or both for MT, (2) presence of granuloma on PB with PF, PB, and negative sputum cultures for fungi, parasite, and atypical mycobacteria and with a clear clinical response to antituberculous treatment, and (3) positive sputum culture for MT associated with exudative pleural effusion, if all other causes of effusion have been ruled out. One hundred and three patients with a definite diagnosis of pleural tuberculosis were finally included. Ten subjects (10%), eight men and two women whose mean age was 29 years (SD=7) were infected with HIV. The remaining 93 patients (90% ), 60 men and 33 women whose mean age was 28 years (SD= 15) had no HIV infection. We found that the HIV -positive group had significantly higher rate of positive sputum cultures for MT (4/ 10 [40%] against 10/ 78 [13%] in noninfected patients, p<0.05), and fewer positive tuberculin skin tests (2/10 [20%], against 57/88 [65%] in HIVnegative patients, p<0.05). In our series, granuloma on pleural CHEST /107161 JUNE, 1995
1775
biopsies, however, were present in only 4 of 9 (44 %) patients infected with HIV, against 72 (84%) in the HIV-negative group, p<0.05, and there were significantly more positive cultures of pleural fluid for MT, 4 of 9 patients (44%) in the HIV-positive group, against 13 (15%) of patients without HIV infection, p<0.05. We conclude that in our study, patients infected with HIV show a lower rate of granulomas on pleural biopsy, however, we agree with Relkin et al that pleural tuberculosis in HIV-positive patients presented itself more often as a manifestation of a greater burden of microorganisms and impaired host response. Nevertheless, as with any retrospective investigation, the results of these studies must be interpreted with caution, and further research is needed to determine the pathophysiologic mechanisms of pleural tuberculosis in patients with HIV infection.
Vicente Gil, MD, Pedro ]. Cordero, MD, jose V. Greses, MD, and juan]. Soler, MD , Hospital Universitari La Fe, Valencia, Spain
REFERENCES 1 Light RW. Pleural diseases. 2nd ed. Philadelphia: Lea & Febiger, 1990; 151-61 2 Jones BE, Young SMM, Antoniskis D, et al. Relationship of the manifestations of tuberculosis to CD4 cell counts in patients with human immunodeficiency virus infection. Am Rev Respir Dis 1993; 148:1292-97
Errata Figure 1 of the transaxial FDG PET image was published upside down in the article by Scott et al, in the April issue of Chest (Chest 1995; 107:1174-76). The legend states that an intense focus of FDG uptake was present in the left supraclavicular fossa. The yellow area in the picture is clearly on the right side of the chest. In Table 1 of "Linkage Analysis of Malignancy-Associated Sarcoidosis" by Reich and colleagues in the March issue of Chest (Chest 1995; 107:605-13), the stubs in column one from top to bottom, were misaligned with the data in the rest of the columns. The corrected Table 1 appears below.
Table !-Summary of Cases of Malignancy Linked With Systemic Granulomata* Patient No. / Age, Yr Race/ Sex
Tumor Type
G-M Interval
Sarcoid Stage
1/ 31 / B/ F
Invasive cervical squamous
5 mo
2/ 69/ W/ M
Fatal lymphoproliferative disorder
None
NA
3/ 59/ W / M
Adenocarcinoma, colon
12 mo
4/ 34/ W/ F
Squamous carcinoma cervix; in situ
2 mo
0 (NCG on-liver biopsy; hypercalcemia; splenomegaly) 0 (fever unknown origin; massive hepatosplenomegally; NCG on liver biopsy specimen)
5/ 36/ W / F
Acute myeloblastic leukemia
Indeterminate; probably simultaneous
6/ 45/ W/ M
Agnogenic myeloid metaplasia
Indeterminate; between 0 and 36mo 1
7/ 34/ W / F
Squamous carcinoma cervix; in situ
Simultaneous
0 (Undifferentiated febrile illness; NCG on liver and lacrimal biopsy specimen)
Evidence of Linkage l. Close temporal proximity 2. Type of malignancy 3. Extensive granulomatous changes in parametrial lymph nodes l. Simultaneous 2. Class of disorder 3. Necrotizing granulomatous changes involving liver, multiple lymph nodes, spleen, and lung l. Close temporal proximity 2. Age at diagnosis uf 5 1;57 years)
1. Close temporal proximity 2. Type of malignancy 3. Unusual pattern of S in a young person l. Simultaneous 2. Type of malignancy 3. Low statistical likelihood of a chance association 4. Smoldering pattern of acute leukemia in a young individual 1. Close temporal proximity 2. Type of malignancy 3. Low statistical likelihood of a chance association l. Simultaneous 2. Type of malignancy 3. Unusual pattern of Sin a young person
*Age=age at diagnosis of malignancy; B=black; W=white; F=female; M=male; G-M interval=time between the recognition of the granulomatous process and the clinical manifestation of malignancy; NA=not applicable; S=sarcoidosis. 'This patient had an unusually low hemoglobin-13.8-at the time sarcoidosis was diagnosed, 3 years before the diagnosis of AMM. 1776
Communications to the Editor
Adults and the Elderly," an article published in Chest in July 1994. They summarized that the young "were more likely to have hemoptysis, fever, and cough ... . " Although the reference probably did not appear in their literature search, there has been a previous description of this phenomenon. The consumption of young men that are in the flower of their age, when the heat of blood is yet brisk, and therefore more disposed to a feverish fermentation, is for the most part acute. But, in old men, where the natural heat is decayed, it is more chronicaJ.l
It is reassuring that some clinical aphorisms have enduring value.
Carl M. Kirsch, MD, FCCP, Division of Respiratory/ Critical Care Medicine , Santa Clara Valley Medical Center, San jose, California
Fistulas Do Not Always Cause Pericardial Effusion
Michael D. Iseman, MD, FCCP, National jewish Center for Immunology and Respiratory Medicine, Denver, Colorado REFERENCE
1 Morton R. Phthisiologia. 1689
To the Editor: Mahaisa variya and colleagues report (Chest 1994; 106:1285-88) on an atrial-esophageal fistula, which adds another interesting case to the roughly 65 comparable cases reported, usually due to benign esophageal disease and sometimes involving the pericardium. The authors note: "The adherence of the esophagus and the myocardium, therefore, provides a potential for fistula development." This needs correcting: the esophagus does not adhere to the myocardium ; the pericardium is interposed between the esophagus and the adjacent cardiovascular structures from the level of T-3 to T-11 1 The probable reason that a fistula does not always cause pericardia! effusion, because it must perforate the interposed pericardium , is the unusual structure of the pericardium over the left atrium-a basketwork of fibers in the parietal pericardium, which tightly clasp the left atrium 2 This is also the reason why none but very large pericardia! effusions under pressure penetrate behind that structure, and why patients stabbed in the back through the left atrium tend to bleed into the pleural cavity rather than the pericardium. These remarks are not made to criticize but rather to amplify a nice report.
David H. Spodick, MD, DSc, Cardiology Division , Saint Vincent Hospital, Worcester, Massachusetts REFERENCES
Spodick DH. Macro- and Micro-physiology and anatomy of the pericardium. Am Heart J 1992; 124:1046-51 2 Spodick DH. The normal and diseased pericardium: current concepts of pericardia! physiology, diagnosis and treatment. J Am Coli Cardiol1983; 1:240-51
Tuberculosis in Young Adults and the Elderly To the Editor: Korzeniewska-Kosela and colleagues (Chest 1994; 106:28-32) reported their observations contrasting "Tuberculosis in Young
Pleural Tuberculosis in HIV-Infected Patients To the Editor: We have read with great interest the article by Relkin et al (Chest 1994; 105:1338-41), which appeared in Chest in May 1994, about the characteristics of pleural tuberculosis in patients infected by HIV. In this report the authors found that patients with pleural tuberculosis and HIV infection, had compared with a group of HIV -negative patients, significantly fewer positive tuberculin skin tests, more acid-fast bacteria identifiable in pleural tissue, higher proportion of positive Mycobacterium tuberculosis (MT) cultures of sputum and pleural biopsy (PB), and similar pleural biopsy histologic conditions. It has been suggested that delayed hypersensitivity plays a major role in the pathogenesis of tuberculous pleural effusion, 1 and T-lymphocytes play a central role in cell-mediated defences against MT. 2 Because HIV infection results in specific depletion of CD4 cells, we might therefore expect to find more acid-fast bacteria identifiable in pleural tissue, sputum, and pleural fluid (PF), and fewer granulomas on PB specimens. To evaluate this hypothesis, we retrospectively studied all cases with a definitive diagnosis of pleural tuberculosis from 1986 through December 1991 at our hospital. Inclusion criteria in the study were (1) positive culture of PF, PB specimen(s), or both for MT, (2) presence of granuloma on PB with PF, PB, and negative sputum cultures for fungi, parasite, and atypical mycobacteria and with a clear clinical response to antituberculous treatment, and (3) positive sputum culture for MT associated with exudative pleural effusion, if all other causes of effusion have been ruled out. One hundred and three patients with a definite diagnosis of pleural tuberculosis were finally included. Ten subjects (10%), eight men and two women whose mean age was 29 years (SD=7) were infected with HIV. The remaining 93 patients (90% ), 60 men and 33 women whose mean age was 28 years (SD= 15) had no HIV infection. We found that the HIV -positive group had significantly higher rate of positive sputum cultures for MT (4/ 10 [40%] against 10/ 78 [13%] in noninfected patients, p<0.05), and fewer positive tuberculin skin tests (2/10 [20%], against 57/88 [65%] in HIVnegative patients, p<0.05). In our series, granuloma on pleural CHEST /107161 JUNE, 1995
1775
biopsies, however, were present in only 4 of 9 (44 %) patients infected with HIV, against 72 (84%) in the HIV-negative group, p<0.05, and there were significantly more positive cultures of pleural fluid for MT, 4 of 9 patients (44%) in the HIV-positive group, against 13 (15%) of patients without HIV infection, p<0.05. We conclude that in our study, patients infected with HIV show a lower rate of granulomas on pleural biopsy, however, we agree with Relkin et al that pleural tuberculosis in HIV-positive patients presented itself more often as a manifestation of a greater burden of microorganisms and impaired host response. Nevertheless, as with any retrospective investigation, the results of these studies must be interpreted with caution, and further research is needed to determine the pathophysiologic mechanisms of pleural tuberculosis in patients with HIV infection.
Vicente Gil, MD, Pedro ]. Cordero, MD, jose V. Greses, MD, and juan]. Soler, MD , Hospital Universitari La Fe, Valencia, Spain
REFERENCES 1 Light RW. Pleural diseases. 2nd ed. Philadelphia: Lea & Febiger, 1990; 151-61 2 Jones BE, Young SMM, Antoniskis D, et al. Relationship of the manifestations of tuberculosis to CD4 cell counts in patients with human immunodeficiency virus infection. Am Rev Respir Dis 1993; 148:1292-97
Errata Figure 1 of the transaxial FDG PET image was published upside down in the article by Scott et al, in the April issue of Chest (Chest 1995; 107:1174-76). The legend states that an intense focus of FDG uptake was present in the left supraclavicular fossa. The yellow area in the picture is clearly on the right side of the chest. In Table 1 of "Linkage Analysis of Malignancy-Associated Sarcoidosis" by Reich and colleagues in the March issue of Chest (Chest 1995; 107:605-13), the stubs in column one from top to bottom, were misaligned with the data in the rest of the columns. The corrected Table 1 appears below.
Table !-Summary of Cases of Malignancy Linked With Systemic Granulomata* Patient No. / Age, Yr Race/ Sex
Tumor Type
G-M Interval
Sarcoid Stage
1/ 31 / B/ F
Invasive cervical squamous
5 mo
2/ 69/ W/ M
Fatal lymphoproliferative disorder
None
NA
3/ 59/ W / M
Adenocarcinoma, colon
12 mo
4/ 34/ W/ F
Squamous carcinoma cervix; in situ
2 mo
0 (NCG on-liver biopsy; hypercalcemia; splenomegaly) 0 (fever unknown origin; massive hepatosplenomegally; NCG on liver biopsy specimen)
5/ 36/ W / F
Acute myeloblastic leukemia
Indeterminate; probably simultaneous
6/ 45/ W/ M
Agnogenic myeloid metaplasia
Indeterminate; between 0 and 36mo 1
7/ 34/ W / F
Squamous carcinoma cervix; in situ
Simultaneous
0 (Undifferentiated febrile illness; NCG on liver and lacrimal biopsy specimen)
Evidence of Linkage l. Close temporal proximity 2. Type of malignancy 3. Extensive granulomatous changes in parametrial lymph nodes l. Simultaneous 2. Class of disorder 3. Necrotizing granulomatous changes involving liver, multiple lymph nodes, spleen, and lung l. Close temporal proximity 2. Age at diagnosis uf 5 1;57 years)
1. Close temporal proximity 2. Type of malignancy 3. Unusual pattern of S in a young person l. Simultaneous 2. Type of malignancy 3. Low statistical likelihood of a chance association 4. Smoldering pattern of acute leukemia in a young individual 1. Close temporal proximity 2. Type of malignancy 3. Low statistical likelihood of a chance association l. Simultaneous 2. Type of malignancy 3. Unusual pattern of Sin a young person
*Age=age at diagnosis of malignancy; B=black; W=white; F=female; M=male; G-M interval=time between the recognition of the granulomatous process and the clinical manifestation of malignancy; NA=not applicable; S=sarcoidosis. 'This patient had an unusually low hemoglobin-13.8-at the time sarcoidosis was diagnosed, 3 years before the diagnosis of AMM. 1776
Communications to the Editor