The importance of the estimation of the quantity and types of airborne microorganisms are that these values can be used as an index for the cleanliness of the environment as well as an index they bear in relation to human health and as source of hospital-acquired infections. The indoor air environment can potentially place patients at greater risk than the outside environment because enclosed spaces can imprison aerosols and allow them to build up to infectious levels (Landrinet al., 2007).

Most of the infections arising from indoor air could potentially be prevented through adequate application of infection control practices (CDC, 2003). For instance, measuring the degree of bacterial contamination of indoor air and the susceptibility pattern of the isolates to commonly used antibiotics in the area will help to select appropriate antibiotics for empirical therapy. This also helps to revise and if necessary, design appropriate hospital infection prevention protocols in an effort to minimize the incidence of costly SSI. Moreover, it provides the tools needed to localize the source and control the spread of SSI (CDC, 2003). Therefore, this study aimed to determine the degree of bacterial contamination of ORs indoor air with respect to acceptable bacterial load standards and measure antimicrobial susceptibility pattern of the isolates.

Study area

This was a descriptive cross-sectional study conducted between November 2014 and January 2015, at Hawassa Referral Hospital, Hawassa, Southern Ethiopia. Hawassa, the capital of Southern Nations, Nationalities, and Peoples' Region (SNNPR) and it is located 275 km from the capital city of Ethiopia, Addis Ababa. The altitude of the town is 1697 km above sea level with mean annual temperature and rainfall of 20.9° o C and 997.6 mm respectively. The total population of Hawassa town is 130,579 with one to one male to female ratio. Hawassa Referral Hospital was inaugurated in November, 2005; it has 850 beds and serves about 12 million people.  The hospital has about 300 daily outpatient visits.

Study design and sampling

A descriptive cross-sectional study was conducted between November 2014 and January 2015 in Hawassa Referral Hospital.

The air samples for the study was taken from six sites; from critical zone (the main surgery OR and gynaecology OR), intermediate zones which comprises female clothing room (FCR)  and  male clothing room (MCR), and from non-critical   zone which include SR (sterilizing room) and  sterilized material store (SMS).

Air samples were collected using  settle  plate or  passive  air  sampling method following 1/1/1 schedule [(a 9cm in diameter sterile Petri dish with 5% Sheep’s blood agar ([9 mm diameter)] was left open to the air for an hour, a meter above the floor and a meter from the wall]) in OR units (Pasquarella et al., 2000). During air sampling sterile gloves, mouth masks and protective gown was worn to prevent self-contamination of the 5% Sheep’s blood agar plate (Oxoid, UK).

The air samples in critical zone of ORs were collected during both active and passive times whereas samples from Intermediate and non-critical zone of ORs was collected in the morning and afternoon in each ward round per week. Then, blood agar plates were transported to the microbiology laboratory and incubated aerobically for 24 hours at 37ºC. The total number of colony forming units (CFU) was enumerated using colony counter and results were expressed in CFU/dm² (Andersson et al., 2012). The colonies were assessed further using the biochemical tests following standard bacteriological techniques to isolate species (WHO, 2010).

The antimicrobial susceptibility testing was done on Mueller-Hinton agar (Oxoid, UK) for every potential pathogenic bacteria isolates with 12 antibiotics each by Kirby-Bauer disk diffusion method matching the test organism to 0.5 McFarland turbidity standards. Then, the susceptibility result was interpreted according to the principles established by Clinical and Laboratory Standards Institute (CLSI) by measuring the zone diameter of inhibition (CLSI, 2005). Reference strains S. aureus (ATCC 25923); E. coli (ATCC 25922) and P. Aeruginosa (ATCC 27853) were used as a quality control for culture and susceptibility testing throughout the study.

All data was analyzed using statistical package for Social Sciences (SPSS) version 16.0 and Microsoft Office Excel. Results were interpreted according to baseline values suggested by Fisher (Fisher et al., 1972; Pasquarella et al., 2000) and values of p<0.05 was considered as statistically significant. The ethical clearance was obtained from Hawassa University ethical review board.

Microbial contamination of air in the OR is commonly well thought-out to be a risk factor for surgical site infections in clean surgery. According to (Pasquarella and colleagues et al., (2000) microbiological quality of air may be considered as mirror of the hygienic condition of the operation theatres. It is also understood that the quality of indoor air depends on internal sources, such as ventilation, cleaning procedures of the surgical team and their activity.

The microbial load in hospital indoor air is highly influenced by the number of occupants, their  activity  and the ventilation. Ventilation causes dilution thus reducing the microbial load. So, indoor bacterial load analysis is very important to know the state of ventilation and cleanness of critical areas within the hospital.

In this study the mean colony counts obtained (Table 1) from all sites except the 93.36 CFU/dm² of gynaecology OT while at active were almost in the acceptable range of bacteriological standard set by Fisher (Pasquarella et al., 2000). Although mean CFU indicates the tolerability range, from the collected 20 samples of each site almost half (10 samples result) shows the unacceptable range. For instance the major OT 50%, gynaecology OT 60% and MCR 60% were under unacceptable level. The possible explanation for the increased load of CFU counts could be the human trafficking particularly students in these room as HURH is a teaching and referral hospital in the region. Besides, the cleaning state of rooms and continues follow-up of sterility of critical areas of the hospital might have offered for the undesirable aerobic bacterial load observed in these rooms. This finding goes in line with the study conducted in Northern Ethiopia (Tewelde et al., 2015) in Jordan (Qudiesat et al., 2009) and also the standard by Fisher (Pasquarella et al., 2000) which explained that the bacterial load of indoor air of OR is higher when it is active than when it is passive. But this studies result was different compared to the findings of the study done in Jimma Ethiopia (Genet et al., 2011) which reported that both ORs were not in the acceptable range.

The mean colony count 85.6 CFU/dm² and 93.4 CFU/dm² obtained from major OT and genecology OT at active time respectively (Table 1) were comparable with a study conducted in Northern Ethiopia (Tewelde et al., 2015) that reported mean colony count of 91.8, 44.9 and 17.2 CFU/dm² for OR respectively.

The highest bacterial populations recorded in the afternoon 27 (45%) samples were beyond the tolerable limit compared to the 38 (63.3%) of the morning samples which were unacceptable compared to standard set by fisher (P=0.045) (Table 3). This result was in line with the finding in Northern Ethiopia (Tewelde et al., 2015b).  But, disagree with Jimma South west Ethiopia finding (Genet et al., 2011) that reported that the colony counts in all OTs except sterilized cloth store were found to be higher in the afternoon than in the morning.

Five bacteria were isolated from a total of 120 air samples examined. Among these S. aureus 54 (45%) and CNS 45 (37.5%) were dominant. In addition, S. aureus was isolated in all sites even though the higher rate was recorded at major OR as a whole (both the critical area along with non-critical area (MCR and FCR). The finding was almost comparable with most of the studies elsewhere, (Shiaka et al., 2014; Qudiesat et al., 2009) which indicates the carrier state of health workers at major OR. On the other hand, the predominant isolates at genecology OR and sterilized material store room was CNS;  the  finding  is  similar  with  studies  conducted   in Jimma (Genet et al., 2011), India (Chacko et al., 2003) and Iraq (Ensayef et al., 2009). The source is more often endogenously from the normal skin flora of patients or exogenously from health workers.

The high level of resistance of S. aureus and CNS to penicillin and tetracycline found in our study has been widely reported throughout the world, (Genet et al., 2011; Tewelde et al., 2015; AlLaham, 2012) and this might be due to emergence of penicillin resistant beta-lactamase producing strains. There was low resistance to amoxcilin, ceftriaxone and vancomycin. The possible explanation for this might be these drugs not intensively used and not easily available outside the hospitals. For P. mirabilies the absence of resistance to amoxyclavulinic in this study make these drugs excellent choices. Ciprofloxacin, trimethoprim-sulphamethoxazole and ceftriaxon were sensitive for to E. coli as well.

In general half of the 20 indoor air samples of major OR were unacceptable bacterial load compared to the standard. As to the entire site, 65% of sampled sites were under the unsatisfactory level of bacterial load. The finding imply that the high bacterial load of indoor air judges it as a risk factor for SSI since there is a linear relationship between bacteria count in air  and surgical site infection rate (Cheesbrough., 2005). On the other hand, the isolates found (S. aureus and CNS) are indicators of the hygiene status of these critical areas.  Hence, to reduce the load of bacterial contamination at critical area as well in the hospital environments, the hospital infection prevention and patient safety (IPPS) should be improved and such type of surveillance should be strengthen.

The authors have not declared any conflict of interests.