Spontaneous and induced sputum values in the diagnosis of pulmonary tuberculosis among patients with suspected pulmonary tuberculosis

Objectives: The quality of sputum samples is important for the diagnosis of pulmonary tuberculosis (TB). Unfortunately, it is not always possible to get a proper sputum sample. Now, sputum induction (SI) has been found to be useful for the diagnosis of many lung diseases. Therefore, this study compares the role of spontaneous sputum and induced sputum in the diagnosis of pulmonary TB among suspected pulmonary TB patients. Materials and Methods: A total of 27 patients admitted with suspected pulmonary TB were studied. Three spontaneous sputum samples were taken from these patients and in case of negative sputum smear microscopy, the SI procedure was used to take another sputum sample. Standard diagnosis included both positive sputum smear and clinical decisions. The sensitivity, specificity, and positive and negative predictive values of sputum smears obtained by two methods of spontaneous sputum and induced sputum were calculated and compared with the standard diagnosis. Results: The result of the induced sputum acid-fast bacillus (AFB) smear microscopy was negative in 16 (59.3%) patients and positive in 11 (40.7%) patients. Six (50%) of those who were negative in spontaneous sputum became positive after SI. The mean white blood cell (WBC) count in normal sputum and induced sputum was 3.8 (±3.4) and 8.8 (±2.9), respectively (P < 0.001). The sensitivity of sputum smear positive was calculated to be 67.7% for SI method and 29.4% for spontaneous sputum method. Conclusion: The diagnostic value of induced sputum in the diagnosis of TB is more than the spontaneous sputum method. In addition, the quality of sputum smear produced through induction is more than spontaneous sputum and it does not create any special complication.

Keywords: Diagnosis, Lung Diseases, Tuberculosis

Tuberculosis (TB) is one of the oldest known diseases. Despite all the success achieved in controlling and preventing infectious diseases in the world and especially in developed countries, TB is still a serious public health problem globally. ^[1],[2],[3]

Delay in the diagnosis of TB is a barrier for the success of TB control program, ^[4],[5] which leads to the spread of this disease in society. Moreover, the diagnosis of TB among human immunodeficiency virus (HIV) infected people is more difficult and it increases the delay in diagnosis. ^[6]

At present, the diagnosis of TB is based on the direct smear microscopy of sputum and its culture; the precise diagnosis is related to the skill of the laboratory personnel and the quality of the sputum sample. ^[6],[7],[8] Nevertheless, the quick and precise diagnosis of smear-positive TB is of great importance ^[9],[10] and it is related to the high-quality sputum sample.

Sputum induction (SI) has been found to be useful in the diagnosis of lung diseases; this is done by collecting the high-quality sputum sample. ^[11],[12] Recently, many studies have been conducted on collecting sputum sample(s) by the SI method from suspected TB patients and significant findings have been collected. ^{[12],[13],[14],[15],[16],[17],[18],[19]} In some studies, the SI results are equal to or better than that of bronchoscopy. ^[20],[21] In this technique, using a device called a nebulizer, hypertonic saline 5% was converted into tiny particles by ultrasound waves; after inhaling these particles, the airways become irritated and it causes coughing. Intra-alveolar osmolality increases and the alveolar fluid moves into the alveoli; the bacilli accumulate and immerse in the fluid. ^[22] This method is noninvasive and less expensive. However, some studies have reported this method to be inefficient in patients with spontaneous sputum. ^{[20],[23],[24]}

So, this study investigated the diagnostic value of SI as a method of collecting sputum samples for the diagnosis of pulmonary TB in patients suspected of pulmonary TB compared with spontaneous sputum.

In this study, all the patients suspected of pulmonary TB who were admitted to the Department of Infectious Diseases in Tohid Hospital (Sanandaj, Kurdistan province, west of Iran) during 2010 were investigated. A total of 27 patients suspected of TB who were hospitalized in the infectious diseases ward were studied. This study was approved by the Ethics Committee of Kurdistan University of Medical Sciences and informed consent was obtained from the participants prior to the study.

Early in the morning, spontaneous sputum samples (three samples) and induced sputum samples after 3 days were obtained from the individuals with suspected pulmonary TB. The patients were instructed to take several deep breaths and then pour their sputum into the cups. In order to adhere to ethics and not to impose an additional diagnosis process on the patients, the first three sputum smears were obtained within 3 days; if the smears were negative, the SI method was used to obtain another sample. To conduct SI, the patients’ mouths were washed prior to SI. Then, the patients were asked to inhale the steam generated by the nebulizer for 20 min through the mask. Steam used in the device contained hypertonic saline 5% and the device output was set on 1 mL/min. This process was interrupted if dyspnea was noted and the patient was under observation for 1 h. When sputum was formed, it was collected in a container and then the sample was sent to a laboratory and a sputum smear was prepared.

Sputum samples were sent to the laboratory. The laboratory personnel had no information about the type of sputum and its owner and all the slides were examined and diagnosed by an expert laboratory specialist. The standard diagnosis was defined as the definitive diagnosis of TB, including having at least two positive smears, or a spontaneous sputum or induced sputum smear technique and a positive culture, or a positive culture and a compatible chest x-ray with pulmonary TB confirmed by a radiologist.­ ^[25] In order to confirm the diagnosis of TB definitively, the patients were followed up for 3 months and were assessed regarding the response to treatment. This procedure was considered to be the gold standard for TB detection.

After reviewing the data, the sensitivity, specificity, and positive and negative predictive values of sputum smears prepared by two methods of spontaneous sputum and induced sputum were calculated and compared with the gold standard diagnosis according to the [Table 1].

Among all the 27 patients, 10 (37%) were healthy and 17 (63%) were smear-positive. Seventeen patients were diagnosed as smear-positive TB patients, out of whom 16 were initially diagnosed as smear-positive and one other person was diagnosed as smear-positive in the later follow-ups.

Among all the suspected cases, 14 (51.9%) were females and 13 (48.1%) were males, 18 (66.7%) were living in urban areas and nine (33.3%) in rural areas. In three cases (11.1%), the subjects had a family history of pulmonary TB [Table 2].

In 24 patients (88.9%), onset of the symptoms took more than 2 weeks. The most common symptom was cough, which was noted in all the patients. Sputum and night sweats were also observed in 25 (92.5%) and 19 (70.4%) patients, respectively. Based on the patients’ chest x-rays, lung infiltration was noted in 12 cases (44.5%) and 10 cases (37%) were normal.

Spontaneous sputum smear microscopy result was negative in 22 (81.5%) patients and positive in five (18.5%). SI smear microscopy result was negative in 16 patients (59.3%) and positive in 11 patients (40.7%). Six patients (50%) who had negative spontaneous sputum smear microscopy became positive sputum after SI. The sensitivity of SI and spontaneous sputum were 67.7% and 29.4%, respectively, [Table 3].

The mean white blood cell (WBC) count in the spontaneous sputum and SI methods was 3.8 (±3.4) and 8.8 (±2.9), respectively, (P < 0.001). The number of normal epithelial cells in the normal sputum and SI methods were 5.7 (±4.5) and 4.1 (±2.2), respectively, (P = 0.14). The patients did not experience any complication during SI.

This study aimed to compare the results of sputum smear collected through two methods, i.e., spontaneous and SI in patients with suspected TB. As mycobacterial culture was not available, the comparison was based on sputum smear microscopy results. The diagnostic value of SI method was more than the spontaneous sputum collection method. In this study, 50% of those with spontaneous smear negative sputum were diagnosed as smear-positive after SI.

Because of the standard diagnosis method in our study that included positivity of smear sputum microscopy through any method and physicians’ decision, the specificity of both the methods were calculated as 100% and we were able to just compare the sensitivity of the methods. Sensitivity was much higher in the SI method. Nonetheless, in this study sensitivity is more important because by the positivity of sputum smear through using any method, the patient was diagnosed with pulmonary TB.

Smear sputum microscopy is the first step of TB diagnosis and it can be negative in 20% to 75% of active TB cases. ^{[12],[15],[16],[17]} Although there are some other diagnostic methods like polymerase chain reaction (PCR) for quicker and more sensitive TB diagnosis, obtaining proper sputum is yet the primary step for utilizing those methods like smear and culture. The positivity of sputum sample is dependent on the quality of the sputum samples. Negativity of the sputum smear can postpone the diagnosis of TB patients. Therefore, methods that improve the quality of the sputum can be helpful in combating TB. SI can increase the quality of sputum samples ^{[11],[12],[18]} and our study also confirmed this. In previous studies, no comparison was made between the number of cells in the sputum obtained through the SI and spontaneous sputum methods. ^[18],[19] In our study, WBC cell counts in SI were higher than the spontaneous method that indicates improvement in the quality of the collected sputum. However, the number of epithelial cells may indicate that in the SI method, a sputum specimen is contaminated with upper respiratory tract secretions. This problem is less encountered in bronchoscopy than in SI.

In some previous studies, the results of sputum smear samples taken by the SI method were even better than bronchoscopy. ^[20] The sixth edition of the “Canada Tuberculosis Standards” recommended using SI instead of bronchoscopy unless all the induced sputum samples are negative or whenever SI is not feasible. ^[26] Therefore, it seems that the SI method can be utilized whenever the presence of TB is highly suspected while smear sputum is negative or the person does not have sputum. ^{[12],[17],[19],[20],[27]} Nevertheless, some studies did not prove this method useful. ^[13],[14] However, in our study, this method does not create any complication for the patients and has no risk associated. Therefore, on following the precautions necessary, performing this procedure is harmless for the patients.

In this method that uses hypertonic saline with or without beta agonists, ^[15] the water steam enters the alveoli and causes the respiratory system to cough and finally generates proper sputum in the lung. Hence, SI not only increases the quantity but also the quality of the sputum. ^[16] Also, this method can be used to study airway inflammation. ^[28]

This study had some limitations. First, the number of patients was low. In addition, it was not possible to use a standard diagnostic method like sputum culture and none of the samples, not even the smear positives, had a positive culture. However, culture still has major problems in developing countries.

According to the results, the diagnostic value of induced sputum in the diagnosis of TB is more than the spontaneous sputum method. In addition, the quality of the sputum smear produced through induction is more than the spontaneous sputum and it does not create any special complication. This method is recommended for usage in the case of patients with suspected TB who are sputum smear-negative or those who do not have sputum.

We would like to thank the Vice Chancellor of Research and Technology, Kurdistan Uni­versity of Medical Sciences, Sanandaj, Kurdistan, Iran who approved this study. This article was part of an MD thesis belonging to Sharmin Sedighi. The authors would like to thank Sharmin Sedighi for collecting the data.

Financial support and sponsorship

Kurdistan University of Medical Sciences.

Conflicts of interest

There are no conflicts of interest.

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/1755-6783.162644

[Table 1], [Table 2], [Table 3]