INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Reactivation pulmonary tuberculosis has been classically considered as a disease causing fibronodular infiltrates, often with cavitation, in one or both upper lung fields, except in older people in whom "atypical" images of lower lung lesions and fewer cavities are common (1, 2). Atypical presentation is supported by studies comparing "young" and "old" tuberculous patients (3). Nevertheless, there is little information about when and how these "atypical" changes occur.

"Atypical" radiologic patterns of pulmonary tuberculosis have also been reported among diabetic patients (4). Most of these reports, however, dealt with a relatively small number of patients, and none of them analyzed the possible influence of age.

The impact of age on the radiologic presentation of pulmonary tuberculosis is important because misinterpretation of the images might delay appropriate diagnostic tests and the start of treatment, thus risking the dissemination of Mycobacterium tuberculosis to others. In order to identify possible age-related changes occurring in pulmonary tuberculosis, the chest radiographs from a large number of tuberculous patients with and without diabetes mellitus were reviewed.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Clinical records from all inpatients admitted to the Instituto Nacional de Enfermedades Respiratorias in Mexico City from 1989 to 1993 with the diagnosis of pulmonary tuberculosis plus diabetes mellitus (TBDM group) were retrospectively reviewed. Of 252 patients found, 51 were excluded because of lack of confirmation of either disease. Thus, we collected 201 patients with active pulmonary tuberculosis confirmed by (1) positive acid-fast bacilli in sputum smear, and/or (2) positive culture for Mycobacterium tuberculosis in sputum, and diabetes mellitus confirmed by at least two fasting serum glucose levels greater than 140 mg/dl. Seven additional patients with miliary tuberculosis (two patients) or pleural tuberculosis (five patients) were excluded, as well as two patients lacking chest radiographs. Therefore, a total of 192 tuberculous patients with diabetes were included in our study. Although a positive sputum culture for Mycobacterium tuberculosis was obtained in only 20% of these patients, according to our clinical bacteriology department, the frequency of atypical mycobacteria in our institute is rather low (10 in 10,000 consecutive positive sputum cultures made from 1993 through 1998). Thus, the chance to include nontuberculous mycobacterial disease in our population was negligible.

The institute where this study was carried out is a national reference center for respiratory diseases. In this institution, the main criteria for hospital admission of tuberculous patients include advanced malnutrition, hemoptysis, respiratory failure, uncontrolled diabetes mellitus, multidrug resistance, pleural effusion, and miliary tuberculosis. However, approximately one third of hospitalized tuberculous patients are admitted for treatment or diagnosis confirmation even in the absence of the above-mentioned criteria. In our studied population there were no patients with risk factors or clinical suspicion of AIDS or HIV infection. In addition, in Mexico there is no clear race distinction as occurs in other countries, and thus racial stratification was not necessary. A control group (TB group) was retrospectively integrated by searching in the institute database for those inpatients discharged between 1990 and 1994 with the ICD 011 ("pulmonary tuberculosis") and without diabetes. From these patients, 248 with bacteriological confirmation (as was described for the diabetic group) were randomly collected. As expected, this population included many tuberculous patients with complications other than diabetes such as pleural and miliary tuberculosis. As we did with the TBDM group, patients with pleural and miliary tuberculosis were excluded, leading to a final sample population of 130 cases.

Every initial posteroanterior chest roentgenogram was reviewed by one pneumologist participating in the study (CPG or ATC). Chest radiographs with difficult images (mainly those with doubtful cavities) were jointly interpreted by both reviewers. According to the site where tuberculous lesions were located, they were classified as upper or lower lung field lesions, when most of them appeared in the upper or lower half, respectively. A cavitation was considered present only when its diameter was larger than 2 cm.

Statistical Analysis

Influence of age and evolution times of tuberculosis and diabetes on some binomial variables (presence or absence of upper lung lesions, lower lung lesions, and cavitations) was assessed through Spearman's correlation coefficient (r_S) by using the proportion of cases in each of seven age groups. Student's t test for unpaired observations was used for some continuous variables. Comparisons between frequency variables were assessed through the chi-square test. Data in the text and illustrations correspond to mean ± SEM (continuous variables) or to frequencies and percentages (noncontinuous variables). Statistical significance was set at two-tailed p < 0.05. Most tests were performed in Epi-Info v6.0 (Centers for Disease Control and Prevention and World Health Organization) and InerSTAT-a v1.3 (Instituto Nacional de Enfermedades Respiratorias, Mexico City).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The male:female ratio in the TBDM group was lower than in the TB group (p < 0.05) (Table 1). In the TB group there were almost twice as many male as female patients, whereas in the TBDM group, there were almost equal number of each. Patients from the TBDM group were significantly older (51.3 ± 0.9 yr; range, 18-90 yr) than patients from the TB group (44.9 ± 1.8 yr; range, 13-90 yr; p < 0.01) (Table 1). The number of patients included in each of the seven age periods in which TBDM and TB groups were divided is also shown in Table 1. Evolution time of tuberculosis was the same (~ 2 yr) in both groups (p = 0.42) (Table 1). This lack of difference was also corroborated when evolution was assessed as time periods (p = 0.84) and when data were adjusted by age periods (data not shown). In addition, TBDM patients had had diabetes for an average of approximately 8 yr.

As can be seen in Figure 1, relationships between radiologic variables and age were analyzed by grouping the patients in seven age-periods. This examination revealed that in the TB group the frequency of any lower lung field lesion steadily increased, from about 30% before age 30, to more than 80% at age 80 or older (r_S = 0.89, p < 0.01) (Figure 1B). The frequency of cavitation in the TB group showed a negative correlation with age, from about 80% at age-periods 30 to 39 and 40 to 49, to less than 20% at age 80 or older (r_S = 0.79 for the complete series, p < 0.05) (Figure 1C). By contrast, the TBDM group showed a high proportion (always more than 70%) of lower lung lesions (r_S = 0.31, p = 0.50) (Figure 1B) or cavitations (r_S = 0.11, p = 0.81) (Figure 1C) at all ages.

View larger version (19K): [in this window] [in a new window]   Figure 1.   Age-induced modifications in the percentage of patients with radiological images showing tuberculous lesions in one or both upper (A) or lower (B) lung fields, or showing one or more cavitations (C ). In the TBDM group, the 80- to 90-yr period was deleted because it was composed of only a 90-yr-old patient with a lower field cavitary lesion.

A steady high proportion (almost always above 80%) of patients with any upper lung lesion was found in both TB (r_S = 0.70, p = 0.077) and TBDM (r_S = 0.06, p = 0.91) groups (Figure 1A). Duration of symptoms of tuberculosis and of diabetes were unrelated to the frequency of lower lung lesion or cavitations (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, we found that the proportion of patients with tuberculosis with lower lung field involvement progressively increased with age, whereas the frequency of cavitation steadily decreased with age. In diabetic patients, lower lung lesions and cavitation were observed in all age groups.

Our data show that the frequency of lower lung involvement by tuberculosis occurs at all ages, but increases with age. It has been suggested that the higher frequency of lower lung disease in the elderly is due to immunologic abnormalities (13) or to a higher frequency of primary tuberculosis (14, 15). We suggest an alternative explanation, based on the concept that the multiplication of Mycobacterium tuberculosis is favored by a high oxygen tension (16). This concept is supported by the finding that reactivation tuberculosis predominantly involves the upper lobes, where PA_O2 is highest (17), and by studies in animal models (18). Aging leads to increased alveolar ventilation and reduced perfusion (21), resulting in increased A/ mismatch and increased P A_O2 (17). These changes should affect the lower lobes more than the upper lobes because the latter have a higher A/ ratio and higher P A_O2. Therefore, age-induced changes should favor multiplication of Mycobacterium tuberculosis in lower lung zones. Furthermore, we found that the frequency of upper lung field lesions (with or without lower lung field lesions) was similar at all ages, suggesting that aging does not alter conditions in the upper lobes.

In diabetics, lower lung lesions were common at all ages, suggesting that diabetes and aging predispose to similar radiologic changes in patients with tuberculosis. Other studies have shown that diabetes causes histologic and functional abnormalities that resemble those found in older people (29).

In patients with tuberculosis alone, cavitation became less common with age, whereas the frequency of cavitation remained high in diabetics of all ages. The increased frequency of cavitation in diabetics was not due to a more prolonged illness as the duration of symptoms prior to diagnosis was similar in diabetics and non-diabetics. The mechanisms by which diabetes alters the frequency of cavitation remain uncertain and require further study.