Article Text


Nasal and conjunctival screening prior to refractive surgery: an observational and cross-sectional study
  1. Koji Kitazawa1,2,3,
  2. Chie Sotozono1,
  3. Masako Sakamoto4,
  4. Miho Sasaki1,
  5. Osamu Hieda1,
  6. Toshihide Yamasaki3,
  7. Shigeru Kinoshita1,2
  1. 1Department of Ophthalmology, Kyoto Prefectural University of Medicine, Kyoto, Japan
  2. 2Department of Frontier Medical Science and Technology for Ophthalmology, Kyoto Prefectural University of Medicine, Kyoto, Japan
  3. 3Baptist Eye Clinic, Kyoto, Japan
  4. 4Research Foundation for Microbial Diseases of Osaka University, Suita, Japan
  1. Correspondence to Dr Koji Kitazawa; kkitazaw{at}


Objectives To investigate bacterial flora of clinically healthy conjunctiva and nasal cavity among patients prior to refractive surgery, as well as the characteristics of patients with methicillin-resistant Staphylococcus aureus (MRSA) colonisation.

Design Observational and cross-sectional study.

Setting A single-centre study in Japan.

Participants 120 consecutive patients pre-refractive surgery.

Primary and secondary outcome measures methods Samples were obtained from the right conjunctival sac and the nasal cavity of 120 consecutive patients prior to refractive surgery and were then measured for the levels of the minimum inhibitory concentration (MIC) of antibiotics. Patients were interviewed regarding their occupation, family living situation and any personal history of atopic dermatitis, asthma, smoking or contact lens wear.

Results Propionibacterium acnes (P. acnes) (32.5%) and Staphylococcus epidermidis (4.2%) were detected from the conjunctival sac. S. epidermidis was the most commonly isolated (68.3%) in the nasal cavity. Of the 30 patients (25.0%) with colonisation by S. aureus, 2 patients, both of whom were healthcare workers with atopic dermatitis, were found to be positive for MRSA in the nasal cavity. A history of contact lens wear, asthma or smoking, as well as patient gender and age, was not associated with MRSA colonisation.

Conclusions There were only 2 patients who were colonised with MRSA, both of whom were healthcare workers with atopic dermatitis. P. acnes was predominantly found in the conjunctival sac. Further study is needed to investigate the involvement between nasal and conjunctival flora, and risk factors for infectious complications.

Statistics from

Strengths and limitations of this study

  • Bacterial flora in the conjunctiva and nasal cavity of the 120 patients before refractive surgery were examined.

  • The characteristics of methicillin-resistant Staphylococcus aureus (MRSA)-positive patients prior to refractive surgery were examined.

  • Small sample size.

  • Relatively low rate of positive bacterial growth from the conjunctival sac.


Although numerous types of refractive surgery, such as laser in situ keratomileusis (LASIK) and photorefractive keratectomy (PRK), are routinely performed worldwide, infectious keratitis postrefractive surgery is a serious disorder which can rapidly lead to loss of vision. The reported incidence of infectious keratitis postrefractive surgery is relatively low, and ranges between 0.011% and 0.091% post-LASIK1–3 and between 0.019% and 0.20% postsurface ablation.3–5 Moreover, the incidence of infectious keratitis postsurface ablation is higher than that of post-LASIK, as epithelial defect and the use of a bandage soft contact lens can increase the risk of infection. Reportedly, fungus, Mycobacterium and Gram-positive bacteria such as methicillin-resistant Staphylococcus aureus (MRSA) are the organisms more likely to cause the infection.6–13 The findings of Chang et al14 showed that Mycobacterium was more likely (47 of 103 infections, 45.6%) than Staphylococcus aureus (S. aureus) (19 of 103 infections, 18.4%) to cause infectious keratitis post-LASIK, while findings by de Rojas et al4 and Donnenfeld et al12 showed that the most likely microorganism to cause infectious keratitis postsurface ablation was Staphylococcus species, including MRSA. Although Mycobacterium is an organism known to be associated with microkeratome, the frequency of infectious keratitis caused by Mycobacterium is reportedly thought to be on the decrease with the rising popularity of femtosecond laser-assisted refractive surgery, due to its high accuracy and low rate of complications.15 Moreover, numerous reports have shown that MRSA-associated keratitis is resistant to multiple drugs and is one of the severe complications that can lead to a sight-threatening infection.2 ,5 ,10 ,12 ,16–22

Chung et al23 reported that in 105 patients scheduled to undergo refractive surgery, coagulase-negative staphylococci (CNS) was isolated from 73 (85%), S. aureus was isolated from 2 (2.3%), Streptococcus pneumoniae was isolated from 1 (1.2%) and Gram-negative bacilli was isolated from 5 (4.8%). In that study, they recommended the use of fourth-generation fluoroquinolones such as gatifloxacin and moxifloxacin, which reportedly have a stronger effect than third-generation fluoroquinolones, such as levofloxacin,24 that are commonly used in Japan to prevent postoperative infection. MRSA is an organism that has resistance to these antibiotics and is known to cause severe postoperative infection. It is necessary to find the high-risk patients for MRSA infection. Although preoperative treatments for ocular surface diseases, as well as povidone-iodine and sufficient eyewash during surgery, can result in a reduction of postoperative infections, it is important to elucidate the patients who are at high risk for MRSA infection.

The purpose of the present study was to investigate bacterial flora in the nasal cavity and conjunctival sac among 120 patients prior to refractive surgery and to clarify the carriage rate of MRSA and its risk factor.


In accordance with the tenets set forth in the Declaration of Helsinki for research involving human subjects, written informed consent was obtained from all subjects prior to their participation in the study.

This study involved 120 consecutive patients (56 males and 64 females; mean age: 33.2±8.8 years (mean±SD) who underwent an initial medical examination for refractive surgery at the Baptist Eye Clinic between February 2011 and May 2012. The associated refractive surgeries included LASIK, photorefractive keratectomy (PRK) and epithelial LASIK (epi-LASIK). Patients with a history of surgical procedures and those using antibiotic eye drops prior to surgery were excluded from the study. Seventeen patients used artificial tears for dry eye. None of the patients with atopic dermatitis were suffering from severe allergic dermatitis or severe ocular disease. Two cases undergoing antiallergy treatment were included; however, both patients used the treatment for mild allergic conjunctivitis and one patient was not suffering from atopic dermatitis.

Samples were obtained from the right conjunctival sac and nasal cavity of each patient via the rotation of a sterile cotton swab on the day of the initial medical examination, and the collected samples were immediately inoculated in culture medium. The samples from the conjunctival sac were inoculated in aerobic and anaerobic culture, and the samples from the nasal cavity were inoculated only in aerobic culture. These samples were transported to the Research Institute for Microbial Disease at Osaka University, Osaka, Japan. The minimum inhibitory concentrations (MICs) of moxifloxacin (MFLX), cefmenoxime (CMX), erythromycin (EM), chloramphenicol (CP), vancomycin (VCM), oxacillin (MPIPC), cefepime (CFPM) and clarithromycin (CAM) were measured for S. aureus. Antibiotics were selected to examine the resistance to multiple drugs, and were not specifically for topical ophthalmic treatment. The fourth-generation fluoroquinolones including MFLX and CMX were basically used as a preoperative and postoperative prophylactic treatment for ocular infection.

On the day of the collection, patients were questioned regarding their occupation, any personal history of atopic dermatitis, asthma and contact lens wear. An additional questionnaire regarding the occupation of the other family members and any history of smoking among the patient and all family members was sent to all of the patients on a different day. Of the 120 patients, 84 answered the additional questionnaire and were analysed.

Statistical analysis

Statistical analysis was performed using JMP software (V.9.1, SAS Institute, Cary, North Carolina, USA). The relationship between colonisation of MRSA and the characteristics of the patients was then statistically evaluated. Fisher's exact test was used to analyse the obtained data.


Of the 120 enrolled patients, bacterial growth was identified in the right conjunctival sac of 45 (37.5%) patients and in the nasal cavity of 99 (82.5%) patients.

Colonisation in the conjunctival sac and nasal cavity

In the conjunctival sac, the following bacteria were found: S. epidermidis (5 patients, 4.2%), Corynebacterium species (1 patient, 0.8%), S. aureus (1 patient, 0.8%), other coagulase-negative staphylococci (CNS) (5 patients, 4.2%) and Propionibacterium acnes (P. acnes) (39 patients, 32.5%) (figure 1). In the nasal cavity, the following bacteria were found: Staphylococcus epidermidis (S. epidermidis) (82 patients, 68.3%), Corynebacterium species (35 patients, 29.2%), S. aureus (30 patients, 25.0%), α-Streptococcus (8 patients, 6.7%) and other CNS (16 patients, 13.3%). In all of the five patients with S. epidermidis in the conjunctival sac, S. epidermidis was observed in the nasal cavity as well.

Figure 1

Isolated bacteria from the nasal cavity (Left) and conjunctival sac (Right). CNS, coagulase-negative staphylococci.

Susceptibility of S. aureus

We examined the antibiotic susceptibility of S. aureus isolated from the conjunctival sac and nasal cavity. S. aureus isolated from the conjunctival sac was methicillin-sensitive S. aureus (MSSA). Among S. aureus isolated from the nasal cavity, 2 strains (6.7%) were MRSA and the other 28 strains were MSSA.

Personal history

To examine the demographic characteristics of the enrolled patients, each patient was questioned in detail about their personal history. The results of the 120 patients showed that this study included 17 patients (14.2%) with atopic dermatitis and 10 patients (22.5%) with asthma. None of the patients with atopic dermatitis had atopic keratoconjunctivitis or severe allergic inflammation on the ocular surface. Twenty patients (16.7%) were healthcare workers, and 105 patients (87.5%) had a past history of contact lens wear.

Characteristics of the patients with MRSA

The demographic characteristics of the patients with or without MRSA in the nasal cavity were examined. No statistically significant differences in age, gender or history of asthma or contact lens wear were found between the patients with and without MRSA colonisation. On the other hand, the carriage rate of MRSA was higher among the healthcare workers (20 patients) than the non-healthcare workers (100 patients) (p<0.05). MRSA was more frequently found in the patients with atopic dermatitis (17 patients) than in those without atopic dermatitis (103 patients) (p<0.05; table 1). The 2 patients with MRSA colonisation were included in 82 additionally questioned patients. The results showed that MRSA was more frequently found in the patients living with healthcare workers (p<0.05), and that a history of smoking had no relation to MRSA colonisation. MIC values against MFLX, CMX, EM, CP, VCM, MPIPC, CFPM, and CAM for MRSA and the summary of the characteristics of the patients with MRSA colonisation are shown in table 2. The results demonstrated that one of the two MRSA isolates showed a high resistance level to multiple drugs. Of the 2 patients with MRSA colonisation, 1 was a doctor with a medical history of atopic dermatitis and whose family member worked as a nurse. The other patient was a nurse with a medical history of atopic dermatitis and who lived with a family member who was a doctor. These results suggest that risk factors for MRSA colonisation included working at a medical institution, a medical history of atopic dermatitis, and living with a family member who works at a medical institution.

Table 1

The demographic characteristics of patients with and without methicillin-resistant Staphylococcus aureus in the nasal cavity

Table 2

The summary of patients with methicillin-resistant Staphylococcus aureus colonisation


In this cross-sectional study, we examined clinically healthy bacterial flora in the nasal cavity and conjunctival sac among 120 patients prior to refractive surgery. The mean patient age prior to refractive surgery was 33.2 years. MRSA was detected in the nasal cavity of two patients who had a history of exposure to the medical treatment environment and a history of atopic dermatitis.

The most common bacteria isolated in the conjunctival sac was P. acnes. CNS was also commonly detected, and the proportion of these bacteria was consistent with that reported in previous studies.23 ,25–27 Chung et al23 reported that CNS was the most prevalent microorganism (85%) isolated from the conjunctival sac of refractive surgery patients, and that 86 (81.9%) of their patients were positive for bacterial growth, while our findings revealed that there was a relatively low rate (37.5%) of positive bacterial growth from the conjunctival sac. However, the longer shipping time, the culture conditions, the culture period and an insufficient rotation of the cotton swab on the conjunctival sac may have contributed to that low positive rate. Although our positive rate of CNS (4.2%) in the conjunctival sac was also low compared to the previous report,23 it may have been a false-negative rate. In addition, the use of bandage contact lenses post surface ablation may have an effect on the conjunctival flora due to bacterial contamination,28 and the differences of postoperative infection needs to be considered between the various surgical procedures.

In the nasal cavity, CNS, Corynebacterium species, and S. aureus were the common bacteria isolated. Previous reports have shown that these bacteria were also major components of the normal nasal flora.29 ,30 Nasal colonisation is a well-known risk factor for subsequent infection.31 ,32 Kimura et al33 also reported a significant relationship between ocular surface colonisation or infection and nasal carriage in patients with MRSA. Actually, S. epidermidis found in the conjunctival sac was observed in the nasal cavity as well. Our findings showed that 2 of the 20 patients who worked as healthcare workers and 2 of the 17 patients with atopic dermatitis were positive for MRSA colonisation in the nasal cavity. Although MRSA was observed in the nasal cavity alone, the nasal screening might be more sensitive than the conjunctival screening.

MRSA is one of the prominent bacteria that are resistant to multiple drugs. MRSA detected in the two patients in this study were also resistant to multiple drugs, but sensitive to MFLX and CP. Salgado et al34 reported a meta-analysis of the prevalence and the risk factors of MRSA colonisation. The findings of their analysis revealed that most MRSA colonisation develops among those who are in contact with other healthcare workers. Reportedly, the prevalence of nasal MRSA colonisation is 1.5–2% in the general population.35–37 Compared to those surveys, our results showed that MRSA-positive rates (10.0%) of nasal flora among healthcare workers (20 patients) were significantly high. Atopic dermatitis is also known to be likely colonised with S. aureus and MRSA (70–80% and 11–23%, respectively),38 ,39 and the higher severity of atopic dermatitis is associated with S. aureus colonisation.40 This result was also consistent with the findings in previously published papers.38 ,39 The use of contact lenses or a history of smoking and asthma have also been reported to affect the bacterial flora,41–43 yet they had no relation to the MRSA colonisation in this study. Solomon et al19 reported that a multicenter chart review of 13 cases of MRSA keratitis post refractive surgery showed that 13 eyes of 12 patients, 9 of whom were either healthcare workers or exposed to a hospital surgical setting, were at risk for developing MRSA keratitis. Nomi et al17 reported two cases of MRSA keratitis post-LASIK surgery, whose respective occupations were a doctor and a nurse. We also experienced one case of MRSA keratitis post-epi-LASIK, and her occupation was a nurse.

It should be noted that the sample size in this study was not large, and that at least 240, 600, 900 or 1700 samples would need to be included, if the MRSA-positive rates in the healthcare workers were 10%, 5%, 4% or 3%, to achieve a conclusion with statistical significance. However, and at the least, we posit that caution should be taken before recommending healthcare workers with atopic dermatitis for refractive surgery, thus highlighting the need for certain measures to be implemented in order to reduce the risk of severe infection. The screening of nasal and conjunctival flora in healthcare workers with atopic dermatitis could help to initiate appropriate treatment, not empirical treatment, as soon as possible when infectious keratitis unfortunately occurs postrefractive surgery.

In conclusion, our findings demonstrated that two patients who had MRSA colonisation in the nasal cavity among 120 patients prior to refractive surgery were both healthcare workers with atopic dermatitis. Further research is needed to investigate the correlation between nasal and conjunctival flora and infectious keratitis postrefractive surgery.


The authors thank John Bush for his excellent help in the preparation of the manuscript, and they also thank Isao Yokota for professional statistician advice.


View Abstract


  • Contributors KK, CS, OH and SK were responsible for conception and design of the study. KK, MS and OH were involved in collection of data. KK, CS, MS, MS and TY were responsible for management of data. KK, CS, MS and MS were responsible for analysis of data. KK, CS and MS interpreted the data. KK and CS were involved in writing of the manuscript. KK, CS, MS, MS, OH, TY and SK approved the final draft of the manuscript. KK and OH provided statistical expertise to the study. TY was responsible for provision of patients. CS and SK were responsible for obtaining funding for the study. KK and CS conducted literature search.

  • Funding This work was supported by an intrahospital research fund from the Baptist Eye Clinic and research funds from the Kyoto Foundation for the Promotion of Medical Science.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval The Institution Review Board of Kyoto Prefectural University of Medicine and Baptist Eye Yamasaki Clinic (approval number: C-860).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement No additional data are available.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.