Sie sind auf Seite 1von 6

ORIGINAL ARTICLE

Relationship between Computed Tomography Findings and Nutritional Status in Elderly Patients with Pulmonary Tuberculosis
Kyoko Okamura 1,4, Nobuhiko Nagata 2, Hiroyuki Kumazoe 3, Satoshi Ikegame 1, Kentaro Wakamatsu 1, Akira Kajiki 1, Yoshinari Kitahara 1, Koichi Takayama 4 and Yoichi Nakanishi 4

Abstract
Objective This study was aimed to identify the relationship between radiographic features of elderly pulmonary TB patients and nutritional deficiency. Methods Ninety-two patients older than 70 years of age with pulmonary TB were retrospectively enrolled. The influence of nutritional parameters, such as serum albumin concentration and peripheral blood total lymphocyte count on CT findings was examined. CT findings of pulmonary TB patients were classified as those including atypical findings (segmental or lobar consolidation in an unusual location, miliary nodules, and hilar and mediastinal lymphadenopathy) or not. The number of segments involved by TB was also counted. Results Age- and gender-adjusted analyses for the nutritional parameters and confounders revealed that hypoalbuminemia, lymphocytopenia, and steroid therapy were significantly related to the presence of atypical CT findings. Furthermore, hypoalbuminemia, lymphocytopenia, and the amount of acid-fast bacilli in sputum smears were significantly related to an increased number of involved segments. In multivariate analysis, only hypoalbuminemia was significantly related to the presence of atypical CT findings (OR: 0.335, 95% CI: 0.142-0.794, p = 0.013) and an increased number of involved segments (OR: 0.145, 95% CI: 0.047-0.453, p = 0.0009). Among the CT findings, the tree-in-bud pattern was the most common in all patients (79.3%). However, the presence of the tree-in-bud pattern was not significantly related to any nutritional parameter. Conclusion Elderly TB patients of poor nutritional status, in particular those with hypoalbuminemia, tended to show atypical CT findings and widespread lesions. The tree-in-bud pulmonary pattern could be observed in TB patients in any nutritional state. Key words: albumin, computed tomography, elderly, nutrition, pulmonary tuberculosis (Intern Med 50: 1809-1814, 2011) (DOI: 10.2169/internalmedicine.50.5198)

Introduction
Early diagnosis and treatment of pulmonary tuberculosis (TB) is critical to disease management. CT has been established as an important tool in the diagnosis of pulmonary TB (1, 2). Although typical findings, such as the tree-in-bud

pattern, single-cavity nodules, and acinar or lobular nodules are well known (1-4), atypical findings sometimes make it difficult to correctly diagnose pulmonary TB. Nutritional depletion is often seen in patients with TB at diagnosis (5, 6). Malnutrition appears to increase the risk of developing TB, particularly in animal models (7). Atypical CT manifestations of pulmonary TB are common in patients

Department of Respiratory Medicine, National Hospital Organization Omuta National Hospital, Japan, Department of Respiratory Medicine, Fukuoka University Chikushi Hospital, Japan, Department of Radiology, National Hospital Organization Omuta National Hospital, Japan and Research Institute for Diseases of the Chest, Graduate School of Medical Sciences, Kyushu University, Japan Received for publication January 22, 2011; Accepted for publication April 27, 2011 Correspondence to Dr. Kyoko Okamura, kyoko-k@kokyu.med.kyushu-u.ac.jp

1809

Intern Med 50: 1809-1814, 2011

DOI: 10.2169/internalmedicine.50.5198

with impaired host immunity (3, 8-10). However, it is not well known whether or not the nutritional state affects the CT appearance of pulmonary TB. Malnutrition is often observed in elderly TB patients. Furthermore, elderly TB patients are reported to present somewhat atypical symptoms (11, 12) or radiological findings (11, 13, 14), and are more likely to die of the disease (11). The proportion of TB patients with lower lung field lesions increases with age, while the frequency of cavitations decreases with age, as observed by CT (11, 13, 14). We hypothesized that malnutrition in elderly TB patients might influence CT findings. The impact of malnutrition on radiologic presentation of pulmonary TB is considered important, because misinterpretation of images might delay the correct diagnosis and the initiation of TB treatment. We therefore aimed to identify the relationship between radiographic features of elderly pulmonary TB patients and nutritional deficiencies.

astinal lymphadenopathy. CT findings were then assigned to two groups: 1) presence of atypical findings or 2) absence of atypical findings. We also counted the number of involved segments, including atypical findings and all other TB-related findings. A pulmonologist (K.O.) and radiologist (H.K.), who were blinded to the nutritional states of patients, analyzed the CT findings independently, with consensus resolution if there was any disagreement. Statistical analysis Data for continuous variables were presented as mean SD. Duration of symptoms, amount of acid-fast bacilli in sputum smears, diabetes mellitus, steroid therapy, gastrectomy, renal disease, and heart disease were considered as potential confounders. Logistic regression was used to test for relationships between each parameter and CT findings. Age- and gender-adjusted odds ratios were estimated for each nutritional parameter and potential confounder. Nutritional parameters with significant age- and gender-adjusted odds ratios were then included in a multivariate logistic regression model with age, gender, and potential confounders, to determine the relationship between nutritional parameters and CT findings. Only potential confounders with significant age- and gender-adjusted odds ratios were included in the analysis. All analyses were performed using a statistical software package (SAS ver.9.2, SAS Institute Inc., Cary, NC, USA). A p-value of <0.05 was considered statistically significant. Responsibility for data K.O. had full access to all data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis.

Methods
Study population Medical records and pre-treatment chest CT images of 92 pulmonary TB patients older than 70 years of age admitted to National Hospital Organization Omuta National Hospital from 2004 to 2008 were reviewed. Only those patients with confirmed positive sputum cultures for Mycobacterium tuberculosis were enrolled. Patients who took anti-TB medication prior to admission or those with only TB pleurisy were excluded. There were no patients with multidrug-resistant TB. This study was approved by the Ethics Board of National Hospital Organization Omuta National Hospital. Informed consent was waived because of the retrospective, observational study design. High ethical standards were maintained in carrying out this study. Study design We retrospectively obtained nutritional parameters, such as serum albumin concentration and peripheral blood total lymphocyte count, from clinical charts. The influence of each parameter on CT findings was examined. To evaluate the relationship between patient status at admission and CT findings, the influence of duration of symptoms (in weeks) before diagnosis, amount of acid-fast bacilli in sputum smears at diagnosis, and 5 comorbidities (diabetes mellitus, steroid therapy, gastrectomy, renal disease, and heart disease) on the CT findings were also analyzed. The amount of acid-fast bacilli in sputum smears was determined using a five-point semi-quantitative scale (-, , 1+, 2+, and 3+) (15). Assessment of CT findings We classified the following CT findings of pulmonary TB patients as atypical: 1) segmental or lobar consolidation in an unusual location (basal segment, middle lobe, lingular and S3 segments, 2) miliary nodules, and 3) hilar and medi-

Results
Patient characteristics Patient characteristics are detailed in Table 1. A total of 92 pulmonary TB patients were included in the study (49 men and 43 women). Mean age of patients was 80.91 5.79 years. Mean serum albumin level and total lymphocyte count were 3.30 0.71 g/dL and 868.50 507.26/mm3, respectively. Mean duration of symptoms before diagnosis was 4.88 weeks. On the semi-quantitative scale, 20, 6, 30, 19, and 17 patients scored -, , 1+, 2+, and 3+ for the amount of acid fast bacilli in sputum smears, respectively. Atypical CT findings and nutritional state Atypical findings were seen in 36 TB patients (39.1%). There were 28 patients (30.4%) with segmental or lobar consolidation in an unusual location (20 patients in basal segments, 11 patients in middle lobes or lingula, and 3 patients in both a basal segment and middle lobe or lingular segment). Eight patients (8.7%) had miliary nodules. One patient (1.1%) had hilar and mediastinal lymphadenopathy.

1810

Intern Med 50: 1809-1814, 2011


Table1.Patient Characteristics
Mean SD or number of patients (%) (n = 92) Age (years) 80.91 5.79 Gender (male/female) 49/43 Comorbidities 40* (43.5) Diabetes mellitus 16 (17.4) Steroid therapy 9 (9.8) Gastrectomy 8 (8.7) Renal disease 6 (6.5) Heart disease 12 (13.0) Duration of symptoms (week) 4.88 Amount of acid-fast bacilli in sputum smears 20 (21.7) 6 (6.5) 1+ 30 (32.6) 2+ 19 (20.7) 3+ 17 (18.5) Serum albumin (g/dL) 3.30 0.71 868.50 507.26 Total lymphocyte count (/mm3) Values are given as mean SD or number of patients (%) unless otherwise indicated. *Total number of patients with comorbidities. Some patients had more than one comorbidity. Variables

DOI: 10.2169/internalmedicine.50.5198

Tree-in-bud pattern and nutritional state The tree-in-bud pattern was observed in 73 patients (79.3%), and it was the most common CT finding. The presence of the tree-in-bud pattern (Table 6) was observed irrespective of the serum albumin level (p=0.501) and total lymphocyte count (p=0.140).

Discussion
The present study demonstrates that atypical CT findings of pulmonary TB and dissemination of the disease in the elderly are found with greater prevalence in malnourished patients, particularly patients with hypoalbuminemia. In Japan, the average age of TB patients is high. Reactivation of latent TB occurs when such patients get older or become malnourished or immunocompromised. Thus, early diagnosis and treatment of TB patients is important for decreasing mortality and preventing the spread of infection. Our findings contribute to the early diagnosis of TB in malnourished elderly patients. There was no relationship between atypical CT findings and the duration of symptoms before diagnosis, or the amount of acid-fast bacilli in sputum smears. Although the amount of acid-fast bacilli in sputum smears had a significant age- and gender-adjusted odds ratio for the presence of !10 involved segments, it was not related to the extent of involved segments in the multivariate analysis. This suggests that malnourished (hypoalbuminemia) TB patients tend to show atypical or disseminated CT findings, regardless of the duration of symptoms or the amount of acid-fast bacilli in sputum smears. Even in the early stages of TB, malnourished patients might show atypical and widely disseminated CT findings. We found no relationship between comorbidities, such as diabetes mellitus, steroid therapy, gastrectomy, renal disease, and heart disease and CT findings. Given that only hypoalbuminemia had a significant odds ratio in the multivariate analysis, nutritional state is considered to be strongly related to atypical CT findings and dissemination. Atypical CT manifestations of pulmonary TB are common in patients with impaired host immunity (3, 8-10). Diabetic and immunocompromised patients have higher prevalence of multiple cavities within any given lesion and nonsegmental distribution in active pulmonary TB compared to those without underlying disease (9). In this study, diabetes and steroid therapy were not significantly related to the presence of atypical CT findings or increased number of involved segments in the multivariate analysis. A number of possible explanations exist for this discrepancy with previous reports. First, this study had only a few patients with diabetes (16 patients) and steroid therapy (9 patients). Second, the definition of atypical CT findings is different among the studies. Third, hypoalbuminemia might be the factor most strongly related to the presence of atypical CT findings. AIDS patients with a CD4 T lymphocyte count of less

One patient had both lobar consolidation in an unusual location and hilar and mediastinal lymphadenopathy. Age- and gender-adjusted analysis for the nutritional parameters revealed that hypoalbuminemia and lymphocytopenia were significantly related to the presence of atypical CT findings (Table 2). Among potential confounders, only steroid therapy had a significant age- and gender-adjusted odds ratio for atypical CT findings. There was no significant relationship between atypical CT findings and duration of symptoms, or the amount of acid-fast bacilli in sputum smears. In the multivariate analysis, however, only hypoalbuminemia was significantly related to the presence of atypical CT findings (OR: 0.335, 95% CI: 0.142-0.794, p=0.013; Table 3). Number of diseased pulmonary segments and nutritional state The average number of involved segments was 10.63. The number of involved segments was divided into two categories, <10 segments (under half of all pulmonary segments) or !10 segments. There were 46 TB patients (50%) with !10 involved segments. Age- and gender-adjusted analysis revealed that hypoalbuminemia and lymphocytopenia were significantly related to the presence of !10 involved segments (Table 4). Among potential confounders, only the amount of acid-fast bacilli in sputum smears had significant age- and gender-adjusted odds ratio for the presence of !10 involved segments. In the multivariate analysis, however, only hypoalbuminemia was significantly related to having !10 involved segments (OR: 0.145, 95% CI: 0.047-0.453, p=0.0009; Table 5).

1811

Intern Med 50: 1809-1814, 2011

DOI: 10.2169/internalmedicine.50.5198

Table2.Rel a tion s hi p between Nutrition a l P a r a meter s a n d Presence of Atypical CT Findings by Age- and Gender- adjusted Analysis
Age- and gender-adjusted OR Serum albumin level 0.247 Total lymphocyte count 0.998 Duration of symptoms 1.085 Amount of acid-fast bacilli 1.255 in sputum smears Diabetes mellitus 1.160 Steroid therapy 6.885 Gastrectomy 0.475 Renal disease 1.276 Heart disease 0.751 OR ,odds ratio 95% CI 0.1130.539 0.9970.999 0.963-1.224 0.908-1.736 0.3603.733 1.30436.336 0.0882.552 0.2336.990 0.2052.747 p value 0.0005 0.004 0.180 0.169 0.804 0.023 0.386 0.779 0.665

Table3.Relationship between Nutritional Parameters and Presence of Atypical CT Findings by Multivariate Analysis
Multivariate 95% CI adjusted OR* Serum albumin level 0.335 0.1420.794 Total lymphocyte count 0.999 0.9981.001 Steroid therapy 4.141 0.716-23.948 *Multivariate adjusted odds ratios (ORs) were estimated from the model which included all factors in the table, age, and gender. p value 0.013 0.281 0.113 logistic regression

Table4.Rel a tion s hi p between Nutrition a l P a r a meter s a n d Presence of 10 Involved Segments by Age- and Gender-adjusted Analysis
Age- and 95% CI p value gender-adjusted OR Serum albumin level 0.111 0.0400.308 < 0.0001 Total lymphocyte count 0.998 0.9970.999 0.002 Duration of symptoms 1.074 0.945-1.221 0.274 Amount of acid-fast bacilli 1.594 1.109-2.293 0.012 in sputum smears Diabetes mellitus 0.630 0.185- 2.151 0.461 Steroid therapy 0.992 0.226- 4.352 0.992 Gastrectomy 1.558 0.321 -7.550 0.582 Renal disease 1.022 0.169-6.194 0.981 Heart disease 0.434 0.108-1.747 0.240 Number of involved segments was categorized into two groups: <10 segments and 10 segments. OR, odds ratio.

than 200/L present with an atypical chest radiographic pattern for reactivation TB (16). HIV-seropositive patients have a lower prevalence of localized parenchymal disease and a higher prevalence of disseminated disease on CT (17). Our results suggest that pulmonary CT findings in malnourished TB patients are similar to those in the immunocompromised state. Such a finding may be attributed to defects in cellmediated immunity associated with malnutrition (3, 7), Malnutrition has a profound effect on cellular immune function, as well as on the immune system of the elderly (18). Therefore, malnutrition may be an important risk factor for developing TB, particularly in the elderly. Atypical radiologic images of pulmonary tuberculosis are reportedly common in the elderly (11, 13, 14). The proportion of TB patients with lower lung field lesions increases

with age, while the frequency of cavitations decreases with age (11, 13, 14). This study found that nutritional state, particularly in elderly TB patients with hypoalbuminemia, influenced the frequency of atypical findings, such as segmental or lobar consolidation in an unusual location or miliary nodules, as well as dissemination of pulmonary diseases. The tree-in-bud pattern has been well established as an important sign for TB diagnosis (3, 4). It was seen in TB patients of all nutritional states in this series. However, atypical findings in TB may mislead a physician and such typical radiologic patterns may be ignored. These data suggest that increased awareness in physicians is necessary in order not to miss typical signs among scattered atypical findings in elderly patients with pulmonary TB. Several methods have been reported for the assessment of

1812

Intern Med 50: 1809-1814, 2011

DOI: 10.2169/internalmedicine.50.5198

Table5.Relationship between Nutritional Parameters and Presence of 10 Involved Segments by Multivariate Analysis
Multivariate 95% CI p value adjusted OR* Serum albumin level 0.145 0.0470.453 0.0009 Total lymphocyte count 1.000 0.9981.001 0.579 Amount of acid-fast bacilli 1.382 0.881-2.168 0.159 in sputum smears Number of involved segments was categorized into two groups: <10 segments and 10 segments. *Multivariate adjusted odds ratios (ORs) were estimated from the logistic regression model which included all factors in the table, age, and gender.

Table6.Relationship between Nutritional Parameters and Presence of the Tree-in-bud Pulmonary Pattern on CT
Age- and gender-adjusted OR 1.284 1.001 95% CI 0.6202.660 1.0001.002 p value 0.501 0.140

Serum albumin level Total lymphocyte count OR, odds ratio

nutritional states of pulmonary TB patients. Kim et al selected four markers (serum albumin, serum cholesterol, BMI, and lymphocyte count) that proved to be good predictors of outcome in miliary TB (19). Onwubalili reported using triceps skinfold thickness, arm muscle circumference, serum albumin, iron, BMI, and total iron binding capacity as nutritional indicators in TB patients (5). However, the measurement of triceps skinfold thickness or arm muscle circumference in TB patients is uncommon. Previous work has established that the outcome of hospitalized TB patients can be predicted based on their nutritional state on admission as assessed by serum albumin concentration, peripheral blood total lymphocyte count, BMI, and PPD induration size (20). In this study, BMI was excluded from the analysis, because the height and body weight of some patients in poor nutritional states could not be measured due to poor general condition on admission. Measurement of serum albumin level is very easy and widely used in the world to evaluate nutritional state. Therefore, we used the serum albumin level as a parameter of nutritional status. In this study, hypoalbuminemia was strongly correlated with atypical CT findings and widespread lesions. The present study has several limitations. First, it is a retrospective study without a control group. A prospective study will be needed to confirm the results of this study. Second, HIV serology was not routinely examined in all patients. In Japan, most patients with AIDS are younger than our study population and live in urban areas. This study was performed in a rural area with a low prevalence of AIDS. Most TB cases in this study were considered to be HIV negative and to involve the reactivation of latent TB. Third, the severity of TB and malnutrition tends to be strongly associated. The present study aimed to elucidate the relationship between nutritional state and CT findings at admission to the hospital. Thus, it is difficult to determine the causeand-effect relationship based on our study design. Fourth,

TB patients often possess other underlying diseases and complications. These comorbidities, such as bacterial pneumonia and chronic cardiac failure, may confound the role of pulmonary TB in producing atypical CT findings. Although the exact cause of atypical CT findings was difficult to identify at times, these findings improved with anti-TB treatment. In conclusion, malnourished elderly TB patients, particularly with hypoalbuminemia, tended to exhibit atypical CT findings and widespread lesions. The tree-in-bud pulmonary pattern could be observed in patients with any nutritional state. It is important not to miss the tree-in-bud pattern which coexists with atypical findings when interpreting chest CTs of elderly pulmonary TB patients with poor nutritional state. Further studies with proper controls will need to be carried out to confirm these results.
Each authors role in the study: Dr. Okamura contributed to the planning and data analysis and was the primary writer of the manuscript. Dr. Nagata contributed to the planning of the study, data analysis, and writing of the manuscript. Drs. Kumazoe, Ikegame, Wakamatsu, Kajiki, Kitahara and Takayama contributed to the data analysis. Dr. Nakanishi contributed to the planning of the study and data analysis. The authors state that they have no Conflict of Interest (COI).

References
1. Webb WR, Muller NL, Naidich DP. Diseases characterized primarily by nodular or reticulonodular opacities. In: High-resolution CT of the Lung. 3rd ed. Lippincott, Williams & Wilkins, Philadelphia, 2001: 315-325. 2. Leung AN. Pulmonary tuberculosis: The essentials. Radiology 210: 307-322, 1999. 3. Jeong YJ, Lee KS. Pulmonary tuberculosis: Up-to-date imaging and management. A J R 191: 834-844, 2008.

1813

Intern Med 50: 1809-1814, 2011

DOI: 10.2169/internalmedicine.50.5198

4. Muller NL, Franquet T, Lee KS, Silva CIS. Imaging of pulmonary infections. In: Pulmonary Tuberculosis. Lippincott, Williams & Wilkins, Philadelphia, 2007: 65-67. 5. Onwubalili JK. Malnutrition among tuberculosis patients in Harrow, England. Eur J Clin Nutr 42: 363-366, 1988. 6. Harries AD, Nkhoma WA, Thompson PJ, Nyangulu DS, Wirima JJ. Nutritional status in Malawian patients with pulmonary tuberculosis and response to chemotherapy. Eur J Clin Nutr 42: 445450, 1988. 7. Cegielski JP, McMurray DN. The relationship between malnutrition and tuberculosis: evidence from studies in humans and experimental animals. Int J Tuberc Lung Dis 8: 286-298, 2004. 8. Yabuuchi H, Murayama S, Murakami J, et al. Correlation of immunologic status with high resolution CT and distributions of pulmonary tuberculosis. Acta Radiologica 43: 44-47, 2002. 9. Ikezoe J, Takeuchi N, Johkoh T, et al. CT appearance of pulmonary tuberculosis in diabetic and immunocompromised patients: Comparison with patients who had no underlying disease. A J R 159: 1175-1179, 1992. 10. Kiyan E, Kilicaslan Z, Gurgan M, Tunaci A, Yildiz A. Clinical and radiographic features of pulmonary tuberculosis in non-AIDS immunocompromised patients. Int J Tuberc Lung Dis 7: 764-770, 2003. 11. Lee JH, Han DH, Song JW, Chung HS. Diagnostic and therapeutic problems of pulmonary tuberculosis in elderly patients. J Korean Med Sci 20: 784-789, 2005. 12. Korzeniewska-Kosela M, Krysl J, Muller N, Black W, Allen E,

13.

14. 15. 16.

17.

18.

19.

20.

FitzGerald JM. Tuberculosis in young adults and the elderly. A prospective comparison study. Chest 106: 28-32, 1994. Perez-Guzman C, Torres-Cruz A, Villarreal-velarde H, Vargas MH. Progressive age-related changes in pulmonary tuberculosis images and the effect of diabetes. Am J Respir Crit Care Med 162: 17381740, 2000. Morris CD. Pulmonary tuberculosis in the elderly: a different disease? Thorax 45: 912-913, 1990. Yamagishi F, Ahiko T, Oguri T, et al. Guideline for Tuberculosis. The Japanese Society for Tuberculosis, Tokyo, 2009: 38-39. Keiper MD, Beumont M, Elshami A, Langlotz CP, Miller WT Jr. CD4 T lymphocyte count and the radiographic presentation of pulmonary tuberculosis. Chest 107: 74-80, 1995. Leung AN, Brauner MW, Gamsu G, et al. Pulmonary tuberculosis: Comparison of CT findings in HIV-seropositive and HIVseronegative patients. Radiology 198: 687-691, 1996. Lesourd BM. Nutrition and immunity in the elderly: modification of immune responses with nutritional treatments. Am J Clin Nutr 66: 478S-484S, 1997. Kim DK, Kim HJ, Kwon SY, et al. Nutritional deficit as a negative prognostic factor in patients with miliary tuberculosis. Eur Respir J 32: 1031-1036, 2008. Nagata N, Matsunaga K, Wakamatsu K, et al. The relation between nutritional state of tuberculous patients admitted in tuberculous ward and their outcome at the time of discharge. Kekkaku 84: 611-616, 2009 (in Japanese, Abstract in English).

2011 The Japanese Society of Internal Medicine http://www.naika.or.jp/imindex.html

1814

Das könnte Ihnen auch gefallen