Bacteriospermia, extended spectrum beta lactamase producing Gram-negative bacteria and other factors associated with male infertility in Mwanza, Tanzania: a need of diagnostic bacteriology for management of male infertility

Background Infections caused by Extended spectrum beta lactamase (ESBL) producing bacterial are global challenge. There is limited information on the magnitude of bacteriospermia, ESBL producing Gram-negative bacteria (GNB) causing bacteriospermia and factors associated with male infertility. This study determined magnitude of bacteriospermia, ESBL-GNB and other factors association with infertility among presumptive infertile men in Mwanza, Tanzania. Methods A cross-sectional hospital-based study was conducted between May 2017 and July 2018 among 137 presumptive infertile men. Semen specimens were self-collected by masturbation into clean, sterile and none-spermicidal containers and processed following laboratory standard operating procedures (SOPs). Data analysis was done using STATA 13.0. Results Gram-negative bacteria were predominantly isolated (86.4%), of which 31.6% were ESBL producers. In a total 44 bacteria were isolated from semen culture. The blaCTX-M gene was detected in 75% of phenotypically confirmed ESBL producers. Infertility was independently found to be associated with abnormal spermatozoa morphology (OR (95%CI): 14.48(3.17–66.05)) and abnormal spermatozoa motility (OR (95%CI): 0.05(0.01–0.24)). However, neither bacteriospermia (OR (95%CI): 0.86(0.29–2.59)) nor ESBL bacteriospermia (OR (95%CI): 0.13(0.01–1.22)) was found to be associated with infertility. Conclusion One third of bacteriospermia is due to ESBL-producers with history of antibiotic use being protective factor for infertility. Abnormal spermatozoa morphology and poor spermatozoa forward motility independently predicted infertility.


Introduction
Bacteria, fungi, protozoa and viruses are agents implicated in male urogenital tract and accessory sex gland infections and account for about 15% of male infertility 1 and causing bacteriospermia 3,4 . Bacteriospermia has been found to cause infertility due to various factors such as: deterioration of spermatogenesis, alteration of acrosome and sperm morphology, auto-immune processes induced by inflammation, increased sperm DNA fragmentation due to formation of reactive oxygen species and obstruction of genital tracts due to fibrosis and inflammation 3,5 .
Production of Extended spectrum beta lactamase (ESBL) among multi-drug resistant (MDR) Gram-negative bacteria is the common antibiotic resistance mechanism 6 . E. coli, K. pneumoniae, Citrobacter spp., Enterobacter spp., Acinetobacter spp. and Pseudomonas aeruginosa are common ESBL producing Gram-negative bacteria (GNB) 6,7 . The blaCTX-M gene is documented to account over 75% of ESBL producing gram negative bacteria clinically isolated however little is known on its prevalence among ESBL producing Gram-negative bacteria causing bacteriospermia 8 . Here in, we report the magnitude of bacteriospermia, blaCTX-M among ESBL producing Gram-negative bacteria and other factors associated with male infertility in Mwanza, Tanzania. This is the first study to report magnitude of bacteriospermia, ESBL producing GNB and other factors associated with male infertility from Tanzania. These data are important in the management of bacteriospermia in our setting where there is high prevalence of ESBL producing Gram-negative bacteria [9][10][11] .

Study design, duration, population and setting
This cross-sectional hospital-based study was conducted between May 2017 and July 2018 involving 137 presumptive infertile men (whose female couples were medically confirmed fertile) attending reproductive health/infertility clinics in Mwanza, Tanzania. Standardized data collection tools used to collect socio-demographic and clinical characteristics of the study participants. Semen specimens were self-collected by masturbation into wide mouth, clean, sterile and spermatozoa non-toxic specimen containers (Hunter Scientific Limited, UK) after consented voluntarily and sexual abstinence of a minimum of 3 days. Patients were instructed to pass urine and then thoroughly cleaned their hands and penis with clean water and non-antiseptic soap 12 . Specimens were brought to Central Pathology Laboratory, department of Histopathology at the Bugando Medical Centre (BMC) for semen analysis at room temperature within 30 minutes after collection and Catholic University of Health and Allied Sciences (CUHAS) multipurpose laboratory in cold box (2-8 0 C) for semen culture within one hour of collection. PCR to detect blaCTX-M gene was done at National Institute of Medical Research (NIMR), Mwanza.
In this study, presumptive male infertility refers to male's inability to make fertile female partner pregnant for a period of ≥ 1 year of active sexual practices without protections while medically confirmed infertility among men refers male infertility [13][14][15] . Male infertility can be medically confirmed by examining spermatozoa quality and quantity in an ejaculate 13,14 . This includes: spermatozoa concentration, morphology and/or forward motility 13,14 .

Semen analysis, culture and identification of significant isolated bacteria
Semen analysis involved the following parameters; colour (grey to opalescent), volume (2-6 ml), viscosity (< 2 cm dropping threads from pipette), pH (7.2-8.2), motility, morphology and spermatozoa count (20-120 million/ milliliter) per SOPs and WHO guidelines12. Semen analysis was performed by skilled and experienced laboratory scientist (> 5 working years) and two other laboratory technicians (> 3 working years) were used to confirm for the validity of the results. Semen specimens were inoculated onto blood agar (BA) and MacConkey agar (MCA) plates followed by aerobic incubation at 37 0 C for 24-48 hours. A pure significant growth (≥ 10 3 CFU/ml growth) of bacteria were further identified to species level by inhouse biochemical identification tests; Gram stain, catalase, slide coagulase, novobiocin, bacitracin, bile esculin and optochin for Gram-positive bacteria and Gram stain, triple sugar iron (TSI), sulfur indole motility (SIM), Simmons citrate, urease and oxidase for Gram-negative bacteria 16 .

Phenotypic detection of ESBL producing gram negative bacteria
Double disc synergy (DDS) technique was used to detect ESBL producing gram negative bacteria as reported previously17. Briefly, ceftazidime-clavulanic acid and ceftazidime plain discs were seeded on MHA plate with test organisms. The plates were incubated for 24 hours at 37 0 C. The difference of zones of inhibitions of ≥ 5mm between ceftazidime-clavulanic acid and ceftazidime plain was interpreted as ESBL producer 18,19 .

Molecular characterization of blaCTX-M gene from ESBL producing gram negative bacteria
Heat treatment technique was performed to extract bacterial DNA with minor modification from previous study 20 . Two colonies of fresh grown bacteria were suspended into DNase/RNase free tubes containing 500 µL of sterile de-ionized water, mixed by vortexing and boiled at 100 0 C for 10 minutes. Tubes were centrifuged at 12000 rpm for 10 minutes to obtain 5 µL of supernatant of each test bacteria for PCR. PCR was performed for phenotypically confirmed ESBL producing gram negative bacteria to determine the presence of blaCTX-M gene as previously reported 21 . CTX-M3G; forward primer: 5'-GTTACAATGTGT-GAGAAGCAG-3' and reverse primer: 5'-CCGTTTC-CGCTATTACAAAC-3' were used. Briefly, PCR amplification was carried on thermocycler machine (GeneAmp® PCR System 9700, ThermoFishers Scientific, Singapore) as previously explained. Briefly, PCR were conditioned at; initial denaturation at 94 0 C for 5 minutes and cycles: 1; denaturation at 94 0 C for 60 seconds, 2; annealing at 55 0 C for 30 seconds and 3; extension at 72 0 C for 60 seconds and final extension at 72 0 C for 5 minutes. PCR products were visualized under UV illumination on gel electrophoresis by using 2% agarose gel stained with redsafe (7.5 µL of redafe were added into 150 ml of TBE suspended with 3g of agarose powder). The amplicon with band size of 1000bp was annotated as blaCTX-M gene, (Figure 1). E. coli ATCC 25922 was used as negative control organisms.

Statistical analysis
Data analysis was done by using STATA 13.0 version. Continuous data were presented as mean (± SD) and categorical data as percentages. Logistic regression analysis was used to show association between male infertility and independent variables. A p-value of less than 0.05 at 95% confidence interval was considered as statistically significant.

Ethical considerations
Ethical clearance to conduct this study was obtained from a joint BMC/CUHAS ethics and review committee and given ethical numbers: CREC 329/2017 and updated in 2018 by certificate number 719/2018. Written informed consent forms were obtained from study participants before enrollment in this study. Laboratory results; semen analysis, and culture and sensitivity were submitted to respective clinicians for patient management

Socio-demographic and clinical characteristics of study participants
A total of 137 participants were enrolled during this study period. The mean age (±SD) and mean infertility duration (±SD) was 33±6.9 years and 2.7±2 years, respectively. The majorities of participants were living in urban areas (64.2%), 54.0% had tertiary education and 97.1% enrolled from BMC. The following participants reported history of; 2.2% fever, 16.1% antibiotic use within one month prior to enrollment in this study, 18.3% UTI and 3.7% sexually transmitted diseases (STDs). Of the 22 participants with history of antibiotic use, 13 used for ≤ 5 days while of 25 participants with history of UTI, 7 purchased antibiotic without prescription, (Table 1).  The mean age (±SD) and mean infertility duration from seeking medical intervention (±SD) of the 15 infertile participants with bacteriospermia was 31.5 (±8.1) years and 2.5 (±1.5) years, respectively. The majority of in-fertile participants with bacteriospermia had semen hyper-viscosity (73.3%, n=11), hypospermia (66.7%, n=10) and poor forward spermatozoa motility (66.7%, n=10) while 4 (26.7%) participants had no spermatozoa in their semen (Table 3).

Discussion
In the current study, male infertility was found among one third of the participants, of which oligospermia was prevalent encountered type followed with azoospermia while one participant had necrospermia as observed previous 22 . Significantly, male infertility was associated with poor forward spermatozoa motility and abnormal spermatozoa morphology. Poor forward spermatozoa motility means that spermatozoa cannot swim properly hence unable to reach the egg for fertilization and abnormal morphology of the spermatozoa means that spermatozoa may be unable to penetrate an egg for fertilization 23,24 . Therefore, the two factors reduce spermatozoa quality and ability of fetilization 25 . These factors may be used as surrogate markers for diagnosis of male infertility 14 .
Neither bacteriospermia nor ESBL bacteriospermia was associated with male infertility as observed elsewhere 3,4 . This might be due to small sample size of this study resulting to wide 95%CI and imprecise estimate of the effect therefore results didn't have statistical significance 26 . However, about one third of infertile men had bacteriospermia, suggesting that bacteria might have adverse impact on spermatozoa quantity and/or quality. This is further supported by the fact that history of antibiotic use was protective factor of male infertility. Therefore, there is a need of infectology to be part of infertility diagnosis and management among presumptive infertile men. Gram-negative bacteria, specifically K. pneumoniae, E. coli and Acinetobacter spp. were prevalently isolated in this study as previously reported from other studies 22,27 . This is contrary to a study3 which reported that, Gram-positive bacteria, specifically Enterococcus faecalis and Staphylococcus aureus were predominantly isolates causing bacteriospermia. This difference may be due to overall increase trend of multi resistant gram negative infections as the commonest cause of bacterial infections, in the current study's setting Gram-negative bacteria are the most leading causative agents of bacterial infections 10 . About one third of the Gram-negative bacteria were found to be ESBL producers with three quarters of phenotypic ESBL producers carrying blaCTX-M gene. The other quarter might be carrying other CTX-M groups and/or other ESBL families (SHV and TEM) as previously observed 28 . It should be noted previously studies 21,29,30 have found blaCTX-M-15 which is a member of group 1 to be the commonest allele (>75%) in the majority of ESBL producers in Tanzania. The observation of about 30% of Gram-negative bacteria from semen culture to carry ESBL genes is significantly higher than what has been observed in other studies 31,32 . This could be due to high ESBL carriage in our setting due to overuse of antibiotics 29,33 . In this study it was observed that 16.1% of participants used antibiotics mainly cephalosporins (40.9%) without prescriptions. As previously observed, in the current study, resistance to non -beta lactam antibiotics was very high among ESBL-producers 33 . The observation is worrisome as treatment options for ESBL producing bacteria are expensive and most of the time not available in most health facilities in developing countries.

Conclusion
The magnitude of bacteriospermia and ESBL bacteriospermia is high among presumptive infertile men. We recommend that, infectology should be part of diagnosis and management of male infertility.

Limitations of the study
Due to limited funds, this study neither characterized other ESBL families (SHV and TEM) nor other CTX-M groups. Furthermore, we did not investigate other pathogens such as fungi and viruses which could have adverse impact on the quality and/or quantity of spermatozoa in semen of infertile men. Another limitation is, we lack data schistosomiasis screening which is endemic in this study setting and reported elsewhere to be associated with male infertility.