Impaired pulmonary function after treatment for tuberculosis: the end of the disease?

OBJECTIVE: : To evaluate the prevalence of pulmonary function abnormalities and to investigate the factors affecting lung function in patients treated for pulmonary tuberculosis.
METHODS: : A total of 214 consecutive patients (132 men and 82 women; 20-82 years of age), treated for pulmonary tuberculosis and followed at a local dispensary, underwent spirometry and plethysmography at least one year after treatment.
RESULTS: : Pulmonary impairment was present in 102 (47.7%) of the 214 patients evaluated. The most common functional alteration was obstructive lung disease (seen in 34.6%). Of the 214 patients, 60 (28.0%) showed reduced pulmonary function (FEV1 below the lower limit of normal). Risk factors for reduced pulmonary function were having had culture-positive pulmonary tuberculosis in the past, being over 50 years of age, having recurrent tuberculosis, and having a lower level of education.
CONCLUSIONS: : Nearly half of all tuberculosis patients evolve to impaired pulmonary function. That underscores the need for pulmonary function testing after the end of treatment. OBJETIVO:: Avaliar a prevalência de alterações da função pulmonar e investigar os fatores que afetam a função pulmonar em pacientes tratados para tuberculose pulmonar. MÉTODOS:: Um total de 214 pacientes consecutivos (132 homens e 82 mulheres; 20-82 anos de idade), tratados para tuberculose pulmonar e acompanhados em um dispensário local, foi submetido a espirometria e pletismografia pelo menos um ano após o tratamento. RESULTADOS:: O comprometimento pulmonar estava presente em 102 (47,7%) dos 214 pacientes avaliados. A alteração funcional mais comum foi o distúrbio ventilatório obstrutivo (observado em 34,6%). Dos 214 pacientes, 60 (28,0%) apresentaram função pulmonar reduzida (VEF1 abaixo do limite inferior de normalidade). Os fatores de risco para função pulmonar reduzida foram tuberculose pulmonar com cultura positiva no passado, idade acima de 50 anos, recidiva de tuberculose e menor nível de escolaridade. CONCLUSÕES:: Quase metade de todos os pacientes com tuberculose evolui com comprometimento da função pulmonar. Isso reforça a necessidade de testes de função pulmonar após o término do tratamento.

INTRODUCTION

It has been estimated that approximately one third of the world's population is latently infected with Mycobacterium tuberculosis . In 2011, there were an estimated 8.7 million new cases of tuberculosis worldwide, equivalent to 125 cases per 100,000 population.

In patients infected with M. tuberculosis , whether treated or untreated, a variety of pulmonary and extrapulmonary sequelae and complications can occur, categorized as follows: parenchymal lesions, which include tuberculoma, thin-walled cavities, scarring, and end-stage lung destruction; or airway lesions, which include bronchiectasis, tracheobronchial stenosis, and broncholithiasis. Structural changes lead to obstructive, restrictive, or mixed patterns of impaired pulmonary function. Studies in patients with pulmonary tuberculosis (PTB) have demonstrated that 33.3-94.0% of such patients develop impaired pulmonary function. Although it is unknown how many PTB survivors are living today, when the incidence of tuberculosis and the success of therapy are considered, the number of PTB survivors appears to be substantial and increasing. There have been few studies on the topic of impaired pulmonary function in PTB survivors, and most such studies have involved highly selected populations. The patients in those populations do not fully represent the populations affected by tuberculosis. - Little is known about the prevalence of a restrictive pattern after PTB. Although the evaluation of true restriction requires the measurement of TLC, the studies cited above used only spirometry. Most of those studies were conducted in countries where the incidence of tuberculosis is low; the prevalence and type of pulmonary impairment might differ in countries where that incidence is high.

The objectives of this study were to evaluate the incidence and extent of pulmonary function abnormalities in patients previously treated for PTB. We also studied the factors affecting pulmonary function in such patients.

METHODS

This study was carried out in the Department of Pulmonary Physiology, Research Institute of Phthisiopulmonology, I.M. Sechenov First Moscow State Medical University, in Moscow, Russia. The study was approved by the Ethical Review Board of the Association of Medical and Pharmaceutical Universities. All participants gave written informed consent, and confidentiality was ensured.

In Russia, the primary component of the tuberculosis control system is the regional dispensary, which provides services to patients with active disease and to those considered at risk of contracting or developing the disease. The patients are followed for some time after treatment.

In the 2003-2007 period, a total of 757 PTB patients (between 20 and 90 years of age) were treated and followed at the local tuberculosis dispensary. All 757 patients were subsequently recruited by telephone. Letters were sent if no telephone contact was made. Of the 757 eligible patients, 214 agreed to take part in the study.

From April 2005 to December 2013, patients previously treated for PTB underwent pulmonary function tests (PFTs) in the Department of Pulmonary Physiology. All 214 of the patients had been treated successfully, and the PFTs were performed at least one year after the end of treatment. At the time of the PFTs, none of the patients showed any signs of active PTB.

All of the participating patients underwent spirometry and plethysmography, as well as completing a questionnaire designed to collect data related to demographic characteristics, smoking, medical history, and other aspects. The PFTs were performed by experienced technicians, in accordance with the recommendations of the American Thoracic Society/European Respiratory Society, , with a combination spirometer/plethysmograph (MasterScreen Body; Jaeger, Würzburg, Germany). Of the 214 patients, 69 had physician-diagnosed chronic lung disease. All 214 patients were urban dwellers without a personal history of exposure to biomass smoke. Those with a history of intermittent bronchodilator use received their usual treatment 30 min prior to testing. European Community for Coal and Steel equations for spirometry and lung volumes were applied. Airway obstruction was defined as an FEV1/VC ratio below the lower limit of normal (LLN) and a TLC above or equal to the LLN; a restrictive pattern was defined as a TLC below the LLN and an FEV1/VC ratio above or equal to the LLN; a mixed pattern was defined as an FEV1/VC ratio and TLC both below the LLN ; and a nonspecific pattern was defined as a TLC above or equal to the LLN, an FEV1/VC ratio above or equal to the LLN, and an FEV1 or FVC below the LLN. We defined reduced pulmonary function as an FEV1 below the LLN.

For quantitative variables, the differences between groups were assessed by the Mann-Whitney test or the Kruskal-Wallis test. For categorical variables, the groups were compared by the chi-square test. The predictive value of age for the presence of pulmonary impairment was evaluated by determining the area under the ROC curve. Accuracy was calculated for the best cut-off value, defined as that with the highest sum of sensitivity and specificity. Logistic regression was used in order to identify factors associated with impaired pulmonary function. Values of p < 0.05 were considered statistically significant.

RESULTS

Table 1 summarizes the main characteristics of the 214 subjects included in the analysis. There were 132 men and 82 women, with a mean age of 51.1 years (range, 20-82 years). Of the 214 subjects, 105 had previously been culture-positive and 90 had previously been culture-negative. For the remaining 19 subjects, the culture results were unknown.

Table 1

Characteristics of the patients in the study sample (n = 214).

Characteristic	Value	Range
Gendera
Male	132 (61.7)
Female	82 (38.3)
Age, yearsb	51.1 (14.2)	20-82
< 40a	46 (21.5)
40-49	59 (27.6)
50-59	49 (22.9)
60-69	32 (15.0)
≥ 70	28 (13.1)
BMI, kg/m2b	23.9 (4.4)	14.8-41.5
Educationa,c
College-educated	66 (31.4)
High school only	144 (68.6)
Smoking statusa
Current smoker	112 (52.3)
Former smoker	26 (12.2)
Never-smoker	76 (35.5)
Pulmonary functionb
FVC, % of predicted	99.7 (21.4)	31-143
FEV1, % of predicted	87.1 (25.0)	23-139
FEV1/VC ratio	70.3 (13.5)	28.7-97
PEF, % of predicted	81.9 (28.7)	21-151
MMEF, % of predicted	61.9 (33.3)	5.4-161
TLC, % of predicted	99.9 (17.8)	38-136
FRC, % of predicted	111.4 (27.9)	45-191
RV, % of predicted	103.9 (30.2)	34-253
IC, % of predicted	91.2 (24.2)	27-165
RV/TLC ratio	34.7 (10.0)	16-74

BMI: body mass index; MMEF: maximal mid-expiratory flow; FRC: functional residual capacity; and IC: inspiratory capacity. aValues expressed as n (%).bValues expressed as mean (SD).cThe level of education was unknown in 4 patients.

The PFT results were normal in 112 (52.4%) of the 214 patients (Table 2). Pulmonary impairment was identified in 102 (47.7%) of the patients, the pattern being obstructive in 74 (34.6%), restrictive in 18 (8.4%), mixed in 8 (3.7%), and nonspecific in 2 (0.9%).

Table 2

Type and severity of pulmonary impairment in patients previously treated for pulmonary tuberculosis.a

Variable	(n = 214)
Type of impairment
Normal	112 (52.4)
Obstructive pattern (FEV1/VC ratio < LLN and TLC ≥ LLN)	74 (34.6)
Restrictive pattern (TLC < LLN and FEV1/VC ratio ≥ LLN)	18 (8.4)
Mixed pattern (TLC and FEV1/VC ratio < LLN)	8 (3.7)
Nonspecific pattern (TLC ≥ LLN, FEV1/VC ratio ≥ LLN, and FEV1 or FVC < LLN)	2 (0.9)
Reduced pulmonary function (FEV1 < LLN)	60 (28)
FEV1
< 35% of predicted	9 (4.2)
35-49% of predicted	11 (5.1)
50-59% of predicted	15 (7.0)
60-69% of predicted	19 (8.9)
≥ 70% of predicted and < LLN	6 (2.8)
≥ LLN	154 (72)

LLN: lower limit of normal. aValues expressed as n (%).

Of the 214 patients, 60 (28.0%) had an FEV1 below the LLN. Using the American Thoracic Society/European Respiratory Society criteria, we classified the degree of abnormality as mild in 6 (2.8%), moderate in 34 (15.9%), and severe in 20 (9.3%), as detailed in Table 2. Clinically significant pulmonary impairment, defined as an FEV1 < 60% of the predicted value, was identified in 35 (14.5%) of the subjects, being attributed to an obstructive pattern in 22 (10.3%), a restrictive pattern in 5 (2.3%), and a mixed pattern in 8 (3.7%). Of those 35 patients, 13 had not been diagnosed with a chronic pulmonary disease.

According to the ROC curve analysis for reduced pulmonary function, when the age of 50 was chosen as the cut-off value, the sensitivity was 73.3% and the specificity was 59.1%. The area under the curve was 0.67 (95% CI: 0.59-0.76; p = 0.002).

As can be seen in Table 3, the prevalence of any pulmonary impairment in the patients ≥ 50 years of age was 60.6% (range, 56.3-63.3%, depending on the specific age bracket), compared with 34.3% (range, 7.1-42.4%, depending on the specific age bracket) in the patients < 50 years of age (p < 0.001; chi-square test). In the patients ≥ 50 of age, the prevalence of reduced pulmonary function was 40.4% (range, 37.5-42.9%, depending on the specific age bracket), whereas it was 15.2% (range, 0.0-20.3%, depending on the specific age bracket) in the patients < 50 years of age, and the difference between the two groups was statistically significant (p < 0.001; chi-square test).

Table 3

Relationship between age and pulmonary impairment in patients previously treated for pulmonary tuberculosis.a

Age, years	Gender	n	Any impairment	Pattern of impairmentb,c			Reduced pulmonary functiond
				Obstructive	Restrictive	Mixed
20-29	Male	7	1 (7.1)	1 (7.1)	0 (0)	0 (0)	0 (0)
	Female	7
	Total	14
30-39	Male	22	10 (31.3)	6 (18.8)	2 (6.3)	1 (3.1)	4 (12.5)
	Female	10
	Total	32
40-49	Male	39	25 (42.4)	24 (40.7)	1 (1.7)	0 (0)	12 (20.3)
	Female	20
	Total	59
50-59	Male	36	31 (63.3)	21 (42.9)	7 (14.3)	3 (6.1)	21 (42.9)
	Female	13
	Total	49
60-69	Male	18	18 (56.3)	13 (40.6)	2 (6.25)	2 (6.3)	12 (37.5)
	Female	14
	Total	32
≥ 70	Male	10	17 (60.7)	9 (32.1)	6 (21.43)	2 (7.1)	11 (39.3)
	Female	18
Total	28

LLN = lower limit of normal. aValues are given as n(%). bAn obstructive pattern was defined as an FEV1/VC ratio < the lower limit of normal (LLN) and a TLC ≥ LLN; a restrictive pattern was defined as a TLC < LLN and an FEV1/VC ratio ≥ LLN; a mixed pattern was defined as an FEV1/VC ratio < LLN and a TLC < LLN; and a nonspecific pattern was defined as a TLC ≥ LLN, an FEV1/VC ratio ≥ LLN, and an FEV1 or FVC < LLN. cTwo patients presented with a nonspecific pattern (data not shown). dDefined as an FEV1 < LLN.

Table 4 shows the pulmonary function parameters, stratified by culture results, number of episodes of PTB, and smoking history. The values for FVC, FEV1, and the FEV1/VC ratio were significantly lower in the patients who had previously had culture-positive PTB than in those who had previously had culture-negative PTB (p < 0.05 for all; Mann-Whitney test). There were 188 patients who had had only one episode of PTB and 26 who had had two or more episodes. Values for FEV1 and the FEV1/VC ratio were significantly lower in the patients who had had two or more episodes of PTB than in those who had had only one. More than 60% of the subjects evaluated in our study had a history of smoking. The prevalence of airway obstruction in ever-smokers (smokers and former smokers, collectively) and never-smokers was 47.1% and 22.4%, respectively (p < 0.001; chi-square test). The prevalence of reduced pulmonary function in ever-smokers and in never-smokers was 31.2% and 22.4%, respectively (p = 0.226; chi-square test). In the smokers, former smokers, and never-smokers, the mean FEV1 (as a percentage of the predicted value) was 87.0 ± 22.7%, 80.0 ± 27.5%, and 89.7 ± 27.1%, respectively (p = 0.201; Kruskal-Wallis test). We found that smoking had no influence on the prevalence of reduced pulmonary function. Only the FEV1/VC ratio was lower in ever-smokers than in never-smokers (Table 4).

Table 4

Conditional analysis of pulmonary impairment after treatment for pulmonary tuberculosis.

Variable	Condition		p
	Previous culture resulta
	Positive	Negative
	(n = 105)	(n = 90)
FVC, % of predictedb	96.7 (22.9)	106.0 (16.8)	0.006
FEV1, % of predictedb	81.8 (24.9)	96.6 (22.0)	< 0.001
FEV1/VC ratiob	67.9 (13.8)	73.6 (12.5)	0.002
TLC, % of predictedb	98.4 (19.4)	103.8 (12.7)	0.106
Any impairmentc	58 (55.2)	30 (33.3)	0.002
Reduced pulmonary functionc	39 (37.1)	13 (14.4)	0.004
	Number of episodes of pulmonary tuberculosis
	Two or more	One
	(n = 26)	(n = 188)
FVC, % of predictedb	90.6 ± 28.9	100.9 ± 20.0	0.140
FEV1, % of predictedb	71.7 ± 27.9	89.2 ± 23.9	0.002
FEV1/VC ratiob	63.3 ± 12.8	71.2 ± 13.4	0.003
TLC, % of predictedb	92.3 ± 25.6	100.9 ± 16.2	0.12
Any impairmentc	20 (76.9)	82 (43.6)	0.001
Reduced pulmonary functionc	15 (57.7)	45 (23.9)	< 0.001
	Smoking
	Ever	Never
	(n = 138)	(n = 76)
FVC, % of predictedb	100.2 ± 20.6	98.7 ± 23.0	0.979
FEV1, % of predictedb	85.6 ± 23.7	89.7 ± 27.1	0.151
FEV1/VC ratiob	67.9 ± 13.3	74.6 ± 12.9	< 0.001
TLC, % of predictedb	102.0 ± 16.0	96.0 ± 20.2	0.072
Any impairmentc	74 (53.6)	28 (36.8)	0.019
Reduced pulmonary functionc	43 (31.2)	17 (22.4)	0.171

Determined for only 195 of the 214 patients in the sample. bValues expressed as mean (SD). cValues expressed as n (%). dDefined as an FEV1 < LLN.

The risk factors for reduced pulmonary function were having previously had culture-positive PTB, being over 50 years of age, having a low level of education, and having experienced recurrence of tuberculosis. The prevalence of reduced pulmonary function was lower in college-educated patients than in those who had only a high school education (18.2% vs. 31.9%), and the difference was statistically significant (p < 0.05; chi-square test). We found that neither gender nor smoking status had any influence on the prevalence of reduced pulmonary function (Table 5).

Table 5

Risk factors for reduced pulmonary function in patients previously treated for pulmonary tuberculosis.a

Risk factor	OR	95% CI	p
Recurrence of tuberculosis: yes vs. no.	4.33	1.86-10.11	< 0.001
Age: < 50 years vs. ≥ 50 years	3.77	1.95-7.25	< 0.001
Positive culture in the pastb: yes vs. no	3.5	1.72-7.11	< 0.001
Education: high school only vs. college education	2.11	1.03-4.33	0.04
Smoking: ever vs. never	1.57	0.82-3.0	0.17
Gender: male vs. female	1.49	0.79-2.81	0.21

Defined as an FEV1 below the lower limit of normal. bDetermined for only 195 of the 214 patients in the sample.

DISCUSSION

A study conducted in the United States showed that 59% of patients treated for tuberculosis subsequently had abnormal pulmonary function. In that study, more than half of the patients treated for PTB evolved to significantly impaired pulmonary function. These data suggest that impaired pulmonary function after PTB is a major cause of chronic lung disease. The authors of that study found that 44% of the patients developed restrictive impairment, compared with the 6.6% reported for the general population. We also found that the prevalence of restrictive impairment was higher among the patients evaluated in our study than in the general population.

In the present study, approximately half of all PTB patients treated at the local tuberculosis dispensary suffered from impaired pulmonary function, a prevalence much higher than that observed in the general population. In addition to the risk factors for reduced pulmonary function identified in the present study-having previously had culture-positive PTB, being over 50 years of age, having a low level of education, and having experienced recurrence of tuberculosis-potential risk factors for impaired pulmonary function in patients previously treated for PTB include extensive disease before treatment, prolonged treatment duration, and poor radiographic improvement after treatment.

We can presume that recurrent PTB is an important risk factor for pulmonary impairment and that patients with risk factors for recurrence require close monitoring and appropriate treatment in order to prevent future episodes of PTB. Hnizdo et al. showed that the mean decrease in FEV1 was 180 mL in patients who had experienced only one episode of PTB, 362 mL in those who had experienced two episodes, and 462 mL in those who had experienced three episodes.

The prevalence of reduced pulmonary function was significantly lower in college-educated patients than in those who had only a high school education; 18.2% and 31.9%, respectively (p < 0.05; chi-square test). As a rule, college-educated patients have a better socioeconomic status, better nutrition, less occupational risk and a more conscientious attitude toward treatment than do those without a college education. The magnitude of the effect of socioeconomic status has been reported to be 200-300 mL of FEV1.

Although smoking has been established as a major risk factor for COPD, an estimated 25-45% of COPD patients are never-smokers and emerging evidence suggests that other risk factors are important. A recent meta-analysis showed that smoking is a risk factor for tuberculosis. However, it is not clear whether smoking can increase the mortality risk in individuals who already have active tuberculosis. The impact of smoking on pulmonary impairment in patients who have been treated for PTB is also unknown. Previous studies have produced inconsistent results. , , , In our study, despite the fact that the prevalence of obstruction was significantly higher in ever-smokers than in never-smokers, we found that smoking had no influence on the prevalence of reduced pulmonary function. The explanation for that finding remains uncertain. Chung et al. found that a history of PTB was a stronger determinant of impaired pulmonary function than was smoking. Those authors also suggested that post-tuberculosis pulmonary inflammation can mask a smoking-related decline in pulmonary function.

The overall prevalence of pulmonary impairment is much higher in patients treated for PTB than in the general population. In our patient sample, obstructive impairment was the most common pattern, followed by restrictive, mixed, and nonspecific impairment. These results are in accordance with those of other studies in the literature, suggesting that pulmonary impairment after tuberculosis could be an underestimated cause of chronic lung disease, especially in countries where the tuberculosis burden is high. Reduced FEV1 has been shown to be an independent predictor of all-cause and respiratory disease-related mortality.

Inghammar et al. showed that impaired pulmonary function was associated with an increased risk of active tuberculosis. However, if decreased lung function is a risk factor for the development of tuberculosis, it is also possible that pulmonary impairment is a risk factor for its recurrence. Further studies are needed in order to determine whether impaired pulmonary function can be viewed as a risk factor and to devise targeted measures to prevent the recurrence of tuberculosis.

The presence of symptoms is not a sensitive and specific indicator of airway limitation, and the use of a symptom questionnaire appears to be an ineffective means of identifying pulmonary impairment. However, evidence suggests that pulmonary function testing is not used consistently. Even in developed countries, less than half of all patients newly diagnosed with chronic pulmonary diseases receive pulmonary function testing near the time of diagnosis. Using a mechanical peak expiratory flow meter or a pocket spirometer as a screening tool can reduce the number of diagnostic PFTs required. Although it has not been established that PFTs can predict recurrence, their use can probably help select a group of patients at higher risk of recurrence who require longer follow-up and prevention measures. This supports the idea that patients previously treated for PTB should undergo pulmonary function testing. However, pulmonary function testing has yet to be included in the guidelines for the treatment of tuberculosis. We suppose that using FEV1 < 80% of the predicted value (rather than FEV1 below the LLN) as a cut-off value would serve as a reasonable strategy for patients under follow-up treatment, because of its simplicity and ease of use.

This study has some limitations. Very few of the patients received bronchodilators, and it is possible that bronchospasm is conducive to pulmonary impairment. In addition, the study was conducted at a single center, whereas a multicenter study could have produced results that would have been more robust. Furthermore, because of a lack of the pertinent data, we did not examine the influence that a history of cavitation had on pulmonary impairment. Tuberculosis treatment is much less effective in patients with cavitary disease than in those with noncavitary disease or culture-positive PTB, and the negative influence of cavitary disease on lung function could therefore be quite significant.

In conclusion, we found that a history of PTB was a risk factor for pulmonary impairment. The eradication of the bacteria does not necessarily translate to the end of the illness. After tuberculosis treatment, more than 40% of patients can evolve to pulmonary impairment, mainly obstructive disorders. This underscores the need for pulmonary function testing in patients who have been treated for PTB.