Lung functions among patients with pulmonary tuberculosis in Dar es Salaam – a cross-sectional study

We conducted a health facility-based descriptive cross sectional study between July and December 2014 in Temeke municipality, the largest of three municipalities in Dar es Salaam city. Two busy TB clinics in this municipality, the Temeke municipal hospital and the Mbagala rangi tatu health centre, were chosen. The target population was patients with pulmonary tuberculosis (PTB), both sputum smear positive and smear negative TB regardless of their HIV status. Eligible patients were aged 18 years or older, had attended the clinic and completed at least 20 of the 24 weeks of anti TB treatment. Prior research has shown that peak lung function loss in patients with TB occurs within 6 months of diagnosis [8]. We enrolled patients who had completed 20 of the 24 weeks of anti TB treatment (i.e. 5 months), since most of these subjects would have achieved a microbiological cure, and that such timing would enable identification of sequel of disease without considering active disease. This timing was also reasonable in allowing data collection for a maximal number of subjects given that patients are discharged from the clinic at 6 months after which it would be difficult to access them. Participants were enrolled in a consecutive manner. The exclusion criteria were participants who had preexisting asthma or COPD, paraplegia, heart failure, stroke, extra pulmonary tuberculosis, contraindications to spirometry (e.g. recent surgery, myocardial infarction, vertigo, vomiting), spine deformities (kyphosis, scoliosis), chest deformities, and pregnancy.

A total of 501 participants were enrolled in the study (Fig. 1). All participants gave consent for participation. Ethical clearance for conducting the study was obtained from the Muhimbili University of Health and Allied Sciences (MUHAS) institutional review board.

Evaluations were incorporated into routine TB clinic visits. A clinical record form (CRF) was used for data collection. A thorough history and physical examination was performed for each participant and recorded on the CRF. The modified Medical research council dyspnea scale was used to assess their functional status [15]. The patients TB02 treatment card was used to obtain other relevant information including duration on anti TB treatment, serostatus for HIV and whether the PTB was smear negative or positive for acid fast bacilli (AFB) at diagnosis. This information was filled in the CRF. A digital weighing scale (OMRON) was used to measure body weight in kilograms (kg). Weight was measured in erect position wearing only light clothing and without shoes. Height was measured in centimeters using meter rods. Waist circumference was measured using a tape measure midway between the lower rib cage margin and the anterior superior iliac spine. Mid upper arm circumference (MUAC) was measured with a tape measure midway between the tip of the shoulder (the acromion) and the tip of the elbow (olecranon process).

This was followed by spirometry by means of the ndd EasyOne™ spirometer (manufactured by ndd Medizintechnik (Switzerland)). The EasyOne™ spirometer complies with the 2005 ATS/ERS spirometry standards. The spirometer does not need daily calibration [16‐18]. It has been used in several studies before both locally and internationally [19‐22]. Thus pre testing of the tool was not performed. The device is portable and not affected by environmental conditions. Precautionary measures in using the spirometer included the use of disposable mouthpiece (spirette) per participant to avoid cross contamination. Spirometry was performed in a vacant area with open sunlight and face masks and gloves were used to protect the investigator/assistants.

The forced expiratory maneuvers, the within and between maneuver evaluation for acceptability as well as test result selection were in accordance with the ATS/ERS guidelines [23]. Spirometry was performed in the standing position. The forced expiratory maneuvers were explained in local language by trained assistants (3 doctors and 1 nurse) to the participants before they underwent spirometry. Spirometry was performed without a nose clip owing to the results of a pilot study among 25 patients prior to this research that observed that patients were uncomfortable when nose clips were used during spirometry. Furthermore, a nose clip is recommended, but not a compulsory requirement for spirometry [23] and other studies have not shown significant differences in results among users and non-users of nose clips [24, 25]. Disposable mouthpieces (spirettes) were discarded after individual use. Parameters measured included forced expiratory volume in one second (FEV1) and the forced vital capacity (FVC). The highest FEV1 and FVC from three acceptable maneuvers were used for further comparison and analysis. The FEV1: FVC% was then calculated from these.

The lung functions were categorized into normal, obstruction, restriction and mixed patterns of lung disease according to the algorithm from the National Lung Health Education Program (NLHEP) [26].

The predicted values were derived from regression equations for healthy Tanzanian adults [22]. To compare the observed and predicted lung functions, we used the percentage of predicted values (the ratio of observed to predicted values times 100) for FVC (FVC %) and FEV1 (FEV1%). Severity for obstruction and restriction were graded according to Global Initiative for Chronic Obstructive Lung Disease (GOLD) [27] and the American Thoracic Society/European Respiratory Society (ATS/ERS) task force recommendations [28] respectively.

Data was analyzed using SPSS (version 15.0) statistical software. Means ± standard deviation (SD) and proportions were calculated for continuous and categorical data respectively. Median and interquartile range (IQR) was used for non-normally distributed numerical data. The Chi square test and t-test were used to determine associations between variables. If more than 20 % of the cells had an expected frequency of <5, Fischer’s exact test was used instead.

There were a total of 501 participants in the study (Fig. 1). Males constituted 60.5 % (303/501) of the study participants. More than 60 % (326/501) of participants were aged 40 years or younger. Majority were married or cohabiting (52.5 %) (263/501), had attained primary level education (65.9 %) (330/501) and were domestic workers (34.1 %) (171/501). Approximately 29.3 % (147/501) were ever smokers with a median of 2.5 pack years. With respect to their clinical characteristics, approximately 54.3 % (272/501) had smear positive pulmonary TB and 90.1 % (245/272) of these had converted to smear negative at 2 months of treatment. Majority (85 %) (426/501) of the patients had presented with a first episode of TB. The median time from first symptom presentation to TB diagnosis was 30 days. Approximately 30 % (152/501) of the population was HIV co-infected. Approximately 46 % (70/152) of the patients had CD4+ cell count <200cells/μL. Majority (81 %) (407/501) had good functional scores (1 to 2) for dyspnea as reported by the MMRC score (Table 1). The mean FEV1, FVC and FEV1: FVC% were 1.83 ± 0.75, 2.95 ± 1.11 and 63.02 ± 17.99 respectively.

The overall prevalence of lung function abnormalities was 74 % (371/501). Majority was due to obstruction (42 %) (210/501) followed by mixed (19 %) (96/501) and restrictive (13 %) (65/501) abnormalities. Approximately 73 % (48/65) of patients with restrictive dysfunction has mild to moderate severity. Approximately 79 % (166/210) of patients with obstructive dysfunction had mild to moderate severity (Fig. 2).

Patients who were aged 40 years or younger were less likely to present with abnormal lung functions (69.9 %) compared to those older than 40 years (81.7 %), p = 0.004. Abnormal lung function was more prevalent among males (78.9 %) than among females (66.7 %) (p = 0.002). However more males developed restrictive (p = 0.034) and mixed (p < 0.001) abnormalities while more females developed obstruction (p < 0.001) (Table 2). As far as clinical characteristics are concerned, higher MMRC scores were significantly associated with any abnormality (84 % for scores 3–5 vs 71.7 % for score 1–2) (p = 0.013). A similar finding was seen for mixed abnormalities (p = 0.009). However lower scores were significantly more associated with obstruction (p = 0.04). Patients who had more than 5 pack years of smoking had significantly higher obstructive abnormalities than those who had smoked for less than 5 pack years (73.3 % vs. 41.9 %) (p =0.005). Patients with recurrent TB were also more likely to have abnormal lung functions compared to those with a first episode of TB (89.3 % vs 71.4 %) (p = 0.001). Also HIV negative patients constituted significantly greater percentage of individuals with abnormal lung functions (77.1 %) compared to HIV infected patients (67.1 %) (p = 0.019). This relationship was also true for mixed lung disease (p = 0.012). However obstructive dysfunction was more common in HIV infected (66.7 %) than in HIV negative patients (52.4 %) (p = 0.014) (Table 3).

In multivariate analysis significant predictors for any abnormality were age > 40 years (Adj OR 1.740, CI 1.080 - 2.804), male sex (Adj OR 1.713, CI 1.123 - 2.614), TB recurence (Adj OR 2.789, CI 1.274 - 6.106) and HIV negative status (Adj OR 1.651, CI 1.055 - 2.583) (Table 4).

Ethical clearance for conducting the study was obtained from the Muhimbili University of Health and Allied Sciences (MUHAS) institutional review board.

Not applicable.

The dataset supporting the conclusions of this article are stored with the Directorate of Research and Publications of the Muhimbili University of Health and Allied Sciences and can be made available upon request.