Sputum Quality Assessment as a Useful Tool for Distinguishing True Pathogensfrom Colonizing Flora

Introduction

The diagnosis of respiratory infections depends on obtaining a high-quality sample. Since sputum cannot always be produced spontaneously, the concept of sputum induction has been in use for several years.¹ When a sputum sample is expectorated, there is a high chance of contamination by oropharyngeal flora in addition to the possible presence of pathogens.² This oropharyngeal flora may outgrow the pathogens in the sputum sample, potentially leading to a false negative result.³ Thus, an expectorated sputum sample is para-mount as it is a quality sample and represents material from the lower respiratory tract.³ In sputum culture, a multitude of species overgrow, making it difficult to identify the actual pathogens.⁴ In order to determine objective criteria for purulence prior to inoculation into culture media, the microbiology laboratory must use the Gram stain screening method.⁵ In the diagnosis, treatment, and outcome of lower respiratory tract infections, the utility of sputum microscopy and culture is however debatable.⁶ Oropharyngeal flora is exclusively distributed in the upper respiratory tract, and the presence of squamous epithelial cells indicate oropharyngeal contamination.⁷ Patients with suspected lung malignancies can also get their sputum cytology test done to check for the presence of malignant cells and different biochemical markers as a prognostic tool during follow-up.⁸ A microscopic evaluation of various cellular components in a Gram stained smear of the specimen under low power field magnification can significantly aid in achieving an accurate diagnosis.² This study was done to assess the quality of sputum samples by Bartlett’s grading system and use it as a tool to distinguish true pathogens from oropharyngeal colonizers.

Materials & Methods

This observational prospective study was conducted over a one-year period in a tertiary care teaching hospital, after obtaining ethical committee clearance (FMIEC/CCM/249/2023). A total of 92 sputum samples were analyzed. The sample size was calculated using the formula, n = z2p(1-p) / e2, Z = 1.96, at 95% confidence interval, P=98, e=2.25% (allowable error).³ Sputum samples from all the adults above 18 years of age, suspected with lower respiratory tract infections were included in the study. Respiratory samples other than sputum and sputum samples from infants and young children below 18 years of age were excluded.

The quality of the expectorated sputum samples received was assessed by the Bartlett’s grading system as shown in Table 1. A final score value of less than or equal to zero indicates salivary contamination of the sputum sample (non-acceptable sputum sample) and a final score of 1 and above was considered as an acceptable sputum sample.

A purulent portion of the sputum sample was selected to do a Gram stain and also to inoculate into appropriate culture media. All bacterial isolates were identified by standard conventional microbiological methods. The antimicrobial susceptibility testing using Modified Kirby Bauer’s disc diffusion method was done and interpreted according to the Clinical Laboratory Standard Institute (CLSI) guidelines, only for the true pathogens.⁹

Results

A total of 92 sputum samples were processed and scored based on Bartlett’s criteria as depicted in Table 1.

Out of 92 samples, 40 (43.4%) were found to contain potential pathogens; 25 (62.5%) of these samples were deemed acceptable while the remaining 15 (37.5%) samples were considered unacceptable as per Bartlett’s criteria. The potential pathogens that were obtained from patients having lower respiratory tract infections (LRTI’s) were Pseudomonas aeruginosa 14 (34.1%), Klebsiella pneumoniae 10 (25%), Acinetobacter baumanii 8 (19.5%), Streptococcus pneumoniae 5 (13%), Escherichia coli 2 (4.9%), and Staphylococcus aureus 1 (2.4%) as shown in Table 2.

The most commonly isolated bacteria causing lower respiratory tract infections were P. aeruginosa 14 (34%), K. pneumoniae 10 (25%), A. baumanii 8(19.50%) and S. pneumoniae 5 (13%) as shown in Table 2. Among the organisms isolated from the unacceptable sputum category, P. aeruginosa was the most frequent 7 (46.6%), followed by A. baumannii 4 (26.6%), K. pneumoniae 3 (20%) and E. coli 1 (6.66%). The clinical presentations among patients with suspected pneumonia included fever, cough with expectoration, breathlessness, hemoptysis and gastrointestinal symptoms. A correlation by Gram stain smear and culture showed significant culture positive isolates 41 (45%) and culture negative ones 51 (55.4%).

Discussion

A specimen of expectorated sputum containing all possible microorganisms poses a diagnostic challenge to the physicians.⁷ The sputum could get contaminated with oropharyngeal flora.² The etiology of bacterial pneumonia has traditionally been studied by examining expectorated sputum; however, the upper respiratory tract flora found in saliva constantly contaminates lower respiratory tract secretions.⁶ High case fatality rates are associated with lower respiratory tract infections, and the random use of antibiotics can alter the oropharyngeal flora, paving the way for the development of multi-drug resistant strains.¹⁰ Thus, physicians are in search of tests that can help them diagnose lower respiratory tract infections, quickly, easily, affordably and conveniently.¹¹ Sputum cultures typically have a higher risk of bacterial growth, with a notable increase in WBC’s and C-reactive protein.¹² The inflammatory process in asthma is mostly caused by eosinophils.¹³ Other obstructive lung disorders, like chronic obstructive pulmonary disease (COPD) or non–cystic fibrosis (CF) bronchiectasis, may have mucus dysregulation as their primary cause.¹⁴ A meta-analysis on Gram stain of high quality sputum samples showed high specificity and positive likelihood ratios (PLR’s) for the diagnosis of S. pneumoniae, H. influenzae, as the causative agents of community acquired pneumonia (CAP).¹⁵ Although it was possible to induce sputum in youngsters, not all of their sputa fit the adult criteria of high quality.¹⁶ In Health Care Associated Pneumonia (HCAP), aspiration pneumonia may occur, leading to a decreased diagnostic yield of sputum Gram stain in these patients compared to community acquired pneumonia (CAP).¹⁷

In our study, out of the 92 samples processed, 53 (57.6%) were acceptable and 39 (42.4%) were non-acceptable, as per the Bartlett’s criteria. This finding however differs from a study by Anuradha et al., where 65% of samples were acceptable, while 9.52% were non-acceptable.⁶ In other studies, such as the one by Mariraj et al., 79% samples were acceptable and 21% were non-acceptable.¹⁰ This implies that there can be a large variation while collecting a sample of sputum and it is paramount that appropriate technique for sample collection be taught at the sample collection facility of a hospital to all the patients. There is always a challenge when a causative bacteria of community acquired pneumonia (CAP) needs to be identified in children, as they can rarely submit a good quality sputum sample. Another hurdle to isolate the desired pathogen in LRTI’s is the initiation of antibiotic therapy before sputum sample collection. All physicians must take note and give instructions accordingly so as to maximize the yield of the respiratory pathogen.¹⁸

Among the potential pathogens obtained in our study, 25 (62.5%) were from acceptable sputum samples and 16 (40%) were from non-acceptable samples as per Bartlett’s criteria. There are bound to be gross variations in different studies in terms of isolation of true pathogens from acceptable samples and other possible pathogens from the non acceptable samples as seen in a study by Renu Goel et al., where true pathogens from acceptable samples were 76.13% and non-acceptable were 23.86%.¹⁹ These variations could be attributed to prior use of antibiotics among these patients, technical errors during processing for culture, or it could be due to incorrect interpretation of the sample growth. The role of potential pathogens from non-acceptable sputum samples is still uncertain. Thus, a clinical evaluation by the treating physician coupled with the culture report is essential to achieve cost effective management of the cases of pneumonia, so as to avoid unscrupulous prescriptions of antibiotics that may lead to increased bacterial resistance and eventually reflect poor antibiotic stewardship.

The potential pathogens identified in a study by Shambhavi et al., were, K. pneumoniae (38.2%), S. pneumoniae (14.8%), P. aeruginosa (14.8%), E. coli (12.7%), S. aureus (8.5%), A. baumanii (8.5%) and Enterococcus spp. (4.2%).²⁰ These pathogens are similar to those obtained in our study and also in a study by Dharmawan et al., who isolated Pseudomonas spp. (31.2%), K. pneumoniae (21.3%), A. baumanii (17.5%), E. coli (15.4%) and S. aureus (5%).²¹ Most of our pathogens were nosocomial in origin barring S. pneumnoiae which was community acquired in nature. Apart from Bartlett’s assessment, Del Rio-Pertuz G et al., observed that sputum quality assessment can also be based on culture and a pneumo CLART technique.²²

Sputum Gram staining is a quick, easy and inexpensive method to assess the quality of a sputum sample. A good staining technique and correct assessment of the morphology of pathogen in relationship to polymorphs by an experienced microbiologist can go a long way in achieving our target. However, poor staining techniques can end up as a burden and cause difficulties in identifying pathogens, increasing the likelihood of prescription of incorrect antibiotics.²³ The total culture positivity in our study was 41 (45%), while a study by Akansha et al., reported higher culture positivity.⁴

At times, patients with underlying cystic fibrosis may experience symptoms such as cough and phlegm, but in our study, the most common symptom among the cases of pneumonia was found to be cough. The other clinical aspects in patients could be fever with cough and expectoration as was seen in a study by Priya M et al.²⁴ The patients may also experience breathlessness apart from cough and fever. The Gram stain smear must be prepared from the most mucopurulent part of the specimen. There should also not be any apprehension that the sample might become diluted if a saline flush is used to induce sputum collection when obtaining a mucopurulent sample is difficult.²⁵

Conclusion

Sputum is an important sample in the diagnosis of lower respiratory tract infections. It is equally important to assess the quality of sputum by a scoring system and process the sample only if it meets the criteria of quality. All salivary samples are best rejected to avoid unnecessary reporting that may lead to prescription of unwarranted antibiotics for an organism that is unlikely to be a pathogen. A good sputum sample alone however is not enough to conclude the existence of LRTI’s, but the report must be correlated with clinical history and examination of the patient. In this study, the most potential pathogens isolated were P. aeruginosa 14 (34.1%), K. pneumoniae 10 (25%), A. baumanii 8 (19.5%), S. pneumoniae 5 (13%), E. coli 2 (4.9%), S. aureus 1 (2.4%) and one fungal isolate Candida albicans 1 (2.4%). This study helped us to judge the quality of sputum and provide a correct report of the potential pathogens from the acceptable sputum samples. The culture report correlated well with the clinical findings and appropriate antibiotics were initiated for the LRTI’s. Thus, it is imperative to use the Bartlett’s scoring or any other scoring system on a routine basis in the microbiology laboratories to achieve correct diagnosis and improve the quality of health care to all the patients.

Conflict of Interest

Nil