Multiplex polymerase chain reaction detection of Streptococcus pneumoniae and Haemophilus influenzae and their antibiotic resistance in patients with community-acquired pneumonia from southwest Iran

Background This study aimed to evaluate the occurrence of Streptococcus pneumoniae and Haemophilus influenzae in sputum of patients with community-acquired pneumonia (CAP) using culture and multiplex polymerase chain reaction (M-PCR) methods and to survey the antibiotic resistance patterns of aforesaid isolates. Result In total, 23.9 % (n = 22/92) of sputum samples showed positive results in the culture method. S. pneumoniae and H. influenzae were isolated from 15 (16.3 %) and 7 (7.6%) samples, respectively. Using M-PCR, 44 (47.8 %) samples were positive for S. pneumoniae and H. influenzae. Of these, S. pneumoniae and H. influenzae were detected in 33 (35.8%) and 11 (11.9%) of the sputum samples, respectively. The sensitivity, specificity, and accuracy rates of PCR in detection of S. pneumoniae in comparison with culture method were 100, 76.6, and 83.6%, respectively. While, the sensitivity, specificity, and accuracy rates of PCR in detection of H. influenzae in comparison with culture method were 100, 95.3, and 95.8%, respectively. Out of 11 isolates of H. influenzae, two strains confirmed as H. influenzae type b (Hib) and 3 isolates were type f. However, 6 isolates were non-typable. The co-trimoxazole and amoxicillin/clavulanate were the less effective antibiotics against S. pneumonia and H. influenzae, respectively. Ceftriaxone with 13.3% resistance rates was the most effective antibiotic against S. pneumoniae, while, clarithromycin, ceftriaxone, and gentamicin with resistance rates of 28.6% for each one were the most effective chemicals against H. influenzae isolates. Conclusion In this study, the prevalence of S. pneumoniae was more than H. influenzae using culture and M-PCR methods. The M-PCR provided better efficiency in detecting the bacterial agents in CAP patients compared to culture method. This method can improve the early detection of pathogens contributed to CAP. The drug resistant S. pneumoniae and H. influenzae indicated the need to develop a codified monitoring program to prevent further spread of these strains.


Background
Over 3 million deaths each year are caused by community-acquired pneumonia (CAP). Also, this infection remains a major challenge to physicians because the highest mortality rates occur in the first days of patient admission [1,2]. Streptococcus pneumoniae and Haemophilus influenzae are two main bacteria that contributed to CAP [3]. Although many pathogens can cause CAP, S. pneumoniae is the most commonly reported bacterium among adults and young children [4]. This Gram-positive pathogen can colonize the nasopharyngeal region without causing significant clinical indications. In developing and underdeveloped nations, S. pneumoniae causes more than 14 million invasive infections and nearly one million mortalities in children each year [5].
H. influenzae is the second most frequently reported bacterium in patients with CAP. This Gram-negative bacterium exists in two forms: encapsulated (typeable) and unencapsulated (non-typeable) H. influenzae (NTHi) [6,7]. Encapsulated H. influenzae includes 6 serotypes a, b, c, d, e and f [8][9][10]. H. influenzae type b (Hib) is a prevalent cause of serious illness, practically in children under the age of 5 years. Before vaccination, the majority of cases of bacterial meningitis in children under 5 years of age were due to Hib, with more than 83% of cases occurring in children under 2 years of age. While the number of cases of Hib has decreased significantly over the last three decades, its impact is still significant. Approximately 340,000 severe cases of Hib infection were reported globally in 2015 among children aged under 5 years old, with most cases (76%) manifesting as pneumonia and 29,600 deaths linked to the Hib infection [9].
As the CAP infection is life-threatening, rapid laboratory diagnosis and immediate treatment is important to control the disease and save the patient life. The culture method is the gold standard in the detection of bacterial pathogens that cause CAP infection. However, it has some limitations. First, it is time-consuming (24-72 h) and the second is low bacterial growth in people who have received antibiotics [11,12]. Hence, using molecular techniques such as polymerase chain reaction (PCR) that are rapid and sensitive, can improve the speed and accuracy of detection of CAP pathogens [13].
Although the antibiotic resistance has increased in most parts of the world and among different bacteria, the true extent of antibiotic non-susceptibility among respiratory pathogens in southwestern Iran has not been well studied. Recent studies from China [14] and Taiwan [15] have reported the emergence of multidrug-resistant S. pneumoniae and H. influenzae strains.
This study aimed to evaluate the occurrence of S. pneumoniae and H. influenzae and their antibiotic resistance patterns in sputum samples of patients with CAP from southwest Iran.

Ethics statement
This study was approved by the Research Ethics Committee (IR.AJUMS.REC.1394.458) of the Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran following the Declaration of Helsinki. Written informed consent was obtained from each patient before collecting sputum samples.

Sample collection
This cross-sectional study was conducted during the two years 2014 to 2016 at the teaching hospitals of Ahvaz city, southwest Iran. The diagnosis of CAP patients was confirmed by specialists in respiratory and lung infectious diseases of the hospitals. Ninety-two sputum samples were collected from patients who were selected based on their clinical symptoms, chest x-ray, and laboratory tests including positive C-reactive protein (CRP), elevated procalcitonin, and leukocytosis. Inclusion criteria were the presence of pneumonia on the basis of a clinical assessment including fever, cough, sputum production, pleuritic chest pain, and dyspnea. Patients who got antibiotic therapy within the last three days were excluded from this study.

Culture and microbiologic testing
The sputum samples were taken in a sterile container and were transported to the laboratory within less than one hour and immediately analyzed microscopically by Gram staining. The samples that had more than 25 polymorphonuclear cells and less than 10 epithelial cells per low power field were included in this study as suitable sputum samples. After Gram staining and direct microscopic examination, sputum samples were cultured on sheep blood agar and chocolate agar (Merck, Germany) and incubated at 35 °C with 5 to 10 % CO 2 . After 48 hours of incubation, the suspected colonies of S. pneumoniae and H. influenzae were identified by a panel of standard biochemical and bacteriological tests. For S. pneumonia, Gram staining, colony morphology, catalase, hemolysis on blood agar plate, bile solubility, and optochin tests were included. For H. influenzae, Gram staining, colony morphology, growth on chocolate agar, oxidase, X factor (hemin) and V factor (nicotinamide-adenine-dinucleotide, NAD) requirement were included [16].

Multiplex polymerase chain reaction (M-PCR) technique
DNA extraction was performed from sputum samples by High Pure PCR Template Preparation Kit (Roche Diagnosis, Mannheim, Germany) according to manufacturers, instructions. The primers used for M-PCR were specific for lytA gene (395bp) of S. pneumoniae and P6 protein (273bp) of H. influenzae (Table 1) [18,19]. The PCR reaction composed of 0.5 μl of each primer, 12.5 μl of Mastermix [50 mM KCl, 10 mmol Tris-HCl (pH 8.3), 0.2 mM each dNTP, 1.5 mM MgCl 2 , 1U Taq DNA polymerase] (Genet Bio, Korea), 5 μl of template DNA, and DNA/RNA free water to reach the final volume of 25 μl. The programs of PCR were as follows: 5 minutes at 95 °C followed by 35 cycles (denaturation at 95 °C for 30 seconds, annealing at 54 °C for 45 seconds and extension at 72 °C for 40 seconds), and the final extension at 72 °C for 7 minutes. S. pneumoniae (ATCC ® 33400 ™ ) and H. influenzae (ATCC ® 33391 ™ ) standard strains were used as positive controls. Also, the E. coli ATCC ® 11775 ™ and master mix without DNA template were used as two negative controls in each PCR run. The electrophoresis (100 V, 40 min) was performed using a 1.5 % agarose gel for the detection of amplified products.

Capsular typing of H. influenzae by PCR
To differentiate encapsulated H. influenzae strains from NTHi strains, the bexA gene was amplified. Capsular typing performed using six specific primer sets which are listed in Table 1 [19,20]. The following standard strains were used for positive control:   450bp [20] each PCR assay were analyzed by gel electrophoresis on 1.5% agarose.

Statistical analysis
The descriptive statistic tests were performed in SPSS version 21.0 (Armonk, NY, USA), and a level of significance of P-value < 0.05 was used. Continuous data was compared using 95 % confidence intervals (CI).

Antimicrobial susceptibility patterns
The antibiotic susceptibility patterns of the S. pneumonia and H. influenzae isolates are shown in Table 3. Ceftriaxone and co-trimoxazole with 13.3 and 73.3% of resistance rates were the most and the less effective antibiotics against S. pneumoniae, respectively. Clarithromycin, ceftriaxone, and gentamicin with resistance rates of 28.6% for each one were the most effective chemicals against H. influenzae isolates. While, the amoxicillin/clavulanate (resistance rate 85.7%) was the weakest drug.

Discussion
Conventional methods, such as culture and serology are not always adequate to detect CAP pathogens. Rapid diagnosis of etiologic agents of CAP based on phenotypic methods is difficult; therefore, new diagnosis methods are needed [21]. It is well known that sensitivity and specificity of routine culture is low for identification of CAP pathogens. Thus, more sensitive and rapid diagnostic methods, such as molecular method, could possibly be useful for detect of CAP pathogens [22]. In this study, both culture and M-PCR methods were used for detection of two bacterial causes (S. pneumoniae and H. influenzae) of CAP. The results showed that the detection rate of M-PCR method (47.8%) was higher than culture method [23.9%]. In line with these findings, Maleki et al. [23] from Iran showed the superiority of molecular techniques such as real-time PCR than culture in detection of S. pneumoniae and H. influenzae from oropharynx and nasal cavity. Also, in another study by et al. [24] from United Kingdom, the detection rates of molecular method were 9.4-26.2% higher than culture for S. pneumoniae, Moraxella catarrhalis, and H. influenzae that was in agreement with this study. Also, in this study the accuracy rate of M-PCR in detection of H. influenzae was higher than S. pneumoniae that was in line with the previous study by Bjarnason et al. [25] from Iceland.
This study revealed a 100% sensitivity of M-PCR in detection of H. influenzae and S. pneumoniae compared to culture method. In contrast to the current research, Gillis et al. [26] showed that PCR based method had poor sensitivity in detection of S. pneumoniae in nasopharyngeal swabs of CAP patients. This discrepancy may be due to the differences in sample types, as we used sputum samples in the current study. The M-PCR method had lower specificity in detection of S. pneumoniae than H. influenzae in this study. Due to the existence of autolysin gene (lytA) in other oral flora streptococci, this crossreactivity may be reduce the M-PCR specificity for detection of S. pneumoniae in sputum samples [25]. Shakib et al. [27] from Iran showed the 100% sensitivity of realtime PCR for detection of S. pneumoniae in sputum samples using the lytA gene that was similar to our results. In another study by Fan et al. [28] from China, a one-step M-PCR assay detecting the ompP6 and the bexA genes of H. influenzae was compared with culture and serum agglutination test. The results of M-PCR showed the sensitivity and specificity of 100 and 99.8%, respectively. This findings were close to our results that exhibited a sensitivity of 100% and the specificity of 95.3% of M-PCR in detection of H. influenzae.
In this study the culture method revealed prevalence rates of 16.3 and 7.6% for S. pneumoniae and H. influenzae, respectively. Using culture method, prevalence rates of 6.5% from Iran [23], 16% from Iceland [25], and 18.4% from China [28] has been reported in previous reports for H. influenzae. Also, occurrence rates of 11.4% from Iran [23] and 20% from Iceland [25] has been reported for S. pneumoniae in previous researches. According to the study of Aydemiret et al. [29] from Turkey, the isolation rates of S. pneumoniae and H. influenzae from sputum, nasopharyngeal swabs and bronchoalveolar lavage (BAL) fluid samples by culture method were 16.2 and 0.0%, respectively. This frequency rate of S. pneumonia was similar to the current study. In contrast to the current study, Costa et al. [6] from Portugal, reported a higher frequency rates of H. influenzae (21.4%) than S. pneumoniae (14.1%) in sputum/BAL samples using culture method.
In the current research, using M-PCR method, the incidence rates of 35.8, 11.9, and 3.2% were revealed These variations among several studies can be due to differences in geographical area, type of detection method, sample size, type of studied sample, and age range of the study population.
Another finding of the current study was the prevalence rates of 2.1 and 6.5% for Hib and NTHi using PCR method, respectively. Also, 3.2% of isolates were H. influenzae type f. Our results were consistent with previous evidence that reported a 2 to 12% prevalence of CAP due to NTHi [10]. Hib is a leading cause of serious illness in children under the age of 5. Previous report from Iran showed the prevalence rate of 8% for Hib in clinical samples [20]. The present study is one of a handful of studies that examined the prevalence of different capsular types of H. influenzae in Iran. In this study, a, c, d, and e types were not observed. In another study from Iran, the b, e, and f capsular types were reported in H. influenzae isolates [33]. There has been evidence from both developed and developing countries that introducing the Hib vaccination has reduced its carrier numbers. In November 2014, the pentavalent vaccine which contains diphtheria, tetanus, whole cell pertussis, hepatitis B, and H. influenzae type b entered the Iranian National Immunization Plan. Vaccination against Hib may reduce Hib colonization, while increasing other serotypes [8,34] The technique used to identify the H. influenzae capsular type may affect the results. Because in some studies, slide agglutination serotyping-based methods and in some others the molecular methods are used [33].
One of the strength of the current study was the determination of antibiotic resistance patterns of S. pneumoniae and H. influenzae isolates against some routine antibiotics. S. pneumoniae showed most susceptibility against ceftriaxone (86.7%) and amoxicillin/ clavulanate (66.7%) which was in agreement with previous studies by Temesgen et al. [32] and Akter et al. [35] who showed the good efficacy of two aforesaid antibiotics against S. pneumoniae. However, the co-trimoxazole with susceptibility rate of 26.7% was the less effective antibiotic against S. pneumoniae. Conversely, Temesgen et al. [32] reported an 8.3% of resistance rate among S. pneumoniae isolates toward the co-trimoxazole. In another study from Iran, the susceptibility rates of S. pneumoniae isolates in patients with non-meningitis invasive disease were 90.4 and 33.3% for ceftriaxone and co-trimoxazole, respectively [36]. Also, the susceptibility rates of S. pneumoniae isolates for erythromycin and clarithromycin were 46.7 and 40%, respectively. The sensitivity rate for erythromycin was higher than a previous report by Houri et al. [36] from Iran who stated the rate of 23.8%, but lower than a report by Temesgen et al. [32] who showed susceptibility rate of 96.7%. However, in a study by Zhao et al. [37] from China, 95.2 and 92.5% of S. pneumoniae isolates were resistant to erythromycin and clarithromycin, respectively. S. pneumoniae is mainly treated with beta-lactams and macrolides, while fluoroquinolones occupied the third treatment choice. There is a major concern in the world regarding beta-lactams and macrolides resistant S. pneumoniae isolates. S. pneumoniae was recently ranked among the 12 bacteria for which new treatments are severely needed by the World Health Organization [38].
The current research showed the good efficacy of clarithromycin and ceftriaxone (more than 70%) against H. influenzae isolates. While, the amoxicillin-clavulanate (resistance rate 85.7%) was the weakest treatment choice. In a previous study by Shooraj et al. [39] from Iran the susceptibility rate of 80% was reported for ceftriaxone that was almost close to this research. Also, our isolates showed the resistance rates of 57.1, and 42.9% for ciprofloxacin and co-trimoxazole, respectively. In contrast to the current study, Shooraj et al. [39] stated the quinolones as the most effective antibiotics against H. influenzae. Also, they showed a higher resistance rate (57.7%) for co-trimoxazole compared to our findings. In another study by Boroumand et al. [40] from Iran, all H. influenza isolates were sensitive to co-trimoxazole. Also, they reported resistance rates of 50 and 45% for ceftriaxone and ciprofloxacin, respectively. In a recent meta-analysis from Iran, the resistance rates of H. influenzae to various antibiotics were as follows: co-trimoxazole 53%, erythromycin 40.3%, ciprofloxacin 30.8%, ceftriaxone 33.1%, and amoxicillin-clavulanate 11.8%. As it is obvious, the rate of amoxicillin-clavulanate resistance in the current study was much higher than the Iranian average. Also, our results were in contrast to previous reported resistance rate for amoxicillin-clavulanate (0.0%) from Ethiopia [32]. Also, in the later study [32], a lower and an equal resistance rates were reported against ciprofloxacin (35.7%) and ceftriaxone (28.6%) against H. influenzae isolates compared to the current research.

Limitations
It is important to acknowledge some limitations of this study. This research could not include all clinical and high risk parameters of patients that may affect the etiology of CAP infection. Also, the focus of this study was based on the prevalence of S. pneumoniae and H. influenzae. Hence, the other bacterial or viral pathogens were not investigated in CAP patients. Also, this study was limited to CAP patients who produced sputum and no blood cultures and urinary antigen test were included. Sputum is still an effective specimen type when investigating CAP microbiologically, despite its shortcomings.

Conclusions
In this study, the M-PCR showed good sensitivity and specificity compared to culture method in detection of H. influenzae and S. pneumoniae in sputum samples of CAP patients. The M-PCR provided better efficiency in detecting the bacterial agents in CAP patients compared to culture method. It is recommended that this technique be used in addition to the culture method in the detection of pathogens involved in the CAP infection. Also, the prevalence of S. pneumoniae was higher than H. influenzae in CAP patients. The Hib was detected in 2.1% of patients. The H. influenzae and S. pneumoniae isolates showed various resistance rates against studied antibiotics. The drug resistant S. pneumoniae and H. influenzae indicated the need to develop a codified monitoring program to prevent further spread of these strains.