Microbiology of acute bacterial dacryocystitis: a tertiary institutional experience in South Australia | International … – Springer

This was a single-centre retrospective study of patients diagnosed with acute dacryocystitis at the Royal Adelaide Hospital over a 5-year period from 2018 to 2023. Patients with incomplete clinical records, no microbiological analysis conducted or those lost to follow-up were excluded. Data was sourced from paper and electronic medical records. Data collected included demographics (age at presentation, sex, past medical history and previous antibiotic use), clinical presentation, radiology findings, microbiological analysis, course of management and clinical outcomes. Microbiological specimens were obtained at first presentation to the ophthalmology clinic. They were obtained by either direct lacrimal aspirate or by wiping a swab across the lower conjunctival cul-de-sac near the medial canthus and punctum by applying pressure over the lacrimal sac area. Purulent material was then transported in Amies medium if conjunctival swab was taken or sterile pot if a lacrimal aspirate was obtained, for culture and sensitivity testing.

Samples were plated and incubated on Columbia horse blood agar (HBA) (5% CO₂ at 35 ± 2 °C for 36 h), Columbia CNA horse blood agar (CAN) (O₂ at 35 ± 2 °C for 36 h), chocolate + Neisseria medium (Choc/NIMM agar) (5% CO₂ at 35 ± 2 °C for 36 h) and anaerobic plain agar plate (anaerobically at 35 ± 2 °C for 84 h) [Edwards group, Narellan, NSW, Australia]. Where significant growth occurred, organisms were identified using matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) mass spectrometry (MALDI Biotyper Sirius IVD system, Bruker Daltonic GMBH, Bremen, Germany). Antimicrobial susceptibility testing (AST) was performed using standard disc diffusion on Mueller Hinton E (MHE) or Mueller Hinton EUCAST agar with horse blood and NAD (MHND) [Edwards group, Narellan, NSW, Australia]. Where applicable, minimum inhibitory concentrations (MICs) were determined by automated microdilution broth testing (BD-Phoenix), Becton Dickinson, Sparks, Maryland, USA). Where applicable, the interpretation of antibiotic susceptibility results was performed according to European Committee on Antimicrobial susceptibility testing (EUCAST) guidelines.

All research was conducted in accordance with the Declaration of Helsinki and was approved by Central Adelaide Local Health Network Human Research Ethics Committee.

This study included 43 patients (Female: 28, Male: 15) between May 2018 and May 2023 (inclusive) from one tertiary oculoplastic unit. The mean age was 63 years (range: 22–90 years). Relevant comorbidities included: diabetes mellitus (9/43, 21%), pre-existing dacryolith (1/43, 2%), nasolacrimal duct stenosis and/or obstruction (17/43, 40%), nasal septal deviation (5/43, 12%), previous trauma to orbit and/or nose (1/43, 2%), neoplasm (2/43, 5%), previous surgery to orbit and/or nose (3/43, 7%). All cases were unilateral. The most common clinical features were: pain (33/43, 77%), swelling (31/43, 72%) and epiphora (24/43, 56%). Other notable clinical features were: discharge (18/43, 42%), localised erythema (12/43, 28%), subjective visual changes (7/43, 16%) and fever (5/43, 12%). 40/43 (93%) patients had a first episode of acute dacryocystitis whilst 3/43 (7%) patients had an episode of acute on chronic dacryocystitis.

Of the 31 cases where swabs or cultures were taken, 4/31 (13%) of cases had a lacrimal aspirate specimen sent, whilst the rest (27/31, 87%) had conjunctival swabs of discharge. Of those who had cultures taken, on microbiological analysis, 21/31 (68%) patients had organisms identified from a lacrimal aspirate or swab, including polymicrobial cultures. No organism was cultured in 10/31 (32%) patients. There was no record of a swab or aspirate being taken for 12/43 (28%) cases. Of the 21 cases with positive cultures, a total of 26 bacterial pathogens were grown. Gram positive organisms comprised 16/26 (62%) of isolates; 9/26 (35%) were Gram negative organisms. Organisms included: Staphylococcus aureus (11/26, 42%), Streptococcus pneumoniae (3/26, 21%), Escherichia coli (2/26, 8%), Aggregatibacter species (2/26, 8%), Streptococcus pyogenes (1/26, 4%), Streptococcus anginosus (1/26, 4%), Morganella morganii (1/26, 4%), Enterobacter cloaceae (1/26, 4%), Hafnia alvei (1/26, 4%), mixed anaerobes (1/26, 4%), E coliforms (1/26, 4%) and Pseudomonas aeruginosa (1/26, 4%). Of the Staphylococcus aureus identified on culture, 2/11 (22%) were methicillin-resistant Staphylococcus aureus (MRSA). A single pathogen was identified in 16/21 (76%) patients, whilst 5/21 (24%) had polymicrobial cultures. Of the five polymicrobial cultures, there was a combination of Streptococcus pneumoniae (1/10, 10%), Streptococcus pyogenes (1/10, 10%), Streptococcus anginosus (1/10, 10%), mixed anaerobes (1/10, 10%), Morganella morganii (1/10, 10%), Aggregatibacter species (1/10, 10%), Hafnia alvei (1/10, 10%) and Staphylococcus aureus (1/10, 10%).

Empirical antibiotics were given in 40/43 (93%) of cases, of which 20/40 (50%) were admitted for a period of intravenous (IV) antibiotics. These regimens included flucloxacillin (12/20, 60%) amoxicillin/clavulanic acid (7/20, 35%), ceftriaxone (5/20, 25%), cefazolin (3/20, 15%), piperacillin/tazobactam (1/20, 5%), metronidazole (1/20, 5%) and vancomycin (1/20, 5%). The remaining 20/40 (50%) who were treated as outpatients were given a course of amoxicillin/clavulanic acid (13/20, 65%), cefalexin (4/20, 20%), azithromycin (1/20, 5%), flucloxacillin (1/20, 5%) and clindamycin (1/20, 5%). The 3/43 (7%) for whom empirical antibiotics were not prescribed, 2/3 (67%) were previously given antibiotics which were not documented in the electronic medical record, 1/3 (33%) was previously on doxycycline long-term and this antibiotic was continued.

Tables 1 and 2 demonstrate the sensitivities of commonly tested antimicrobials in the Gram positive (Table 1) and Gram negative (Table 2) organisms identified in this study. Gram positive organisms were all sensitive to vancomycin, TMP/SXT and chloramphenicol. 93% of Gram positive organisms on which Clindamycin was tested (Staphylococcus aureus, Streptococcus pneumoniae, Streptococcus pyogenes) were sensitive. 83% Gram positive organisms on which cefalexin were tested (Staphyloccocus aureus, Streptococcus pyogenes) were sensitive. 82% of Staphylococcus aureus organisms cultured were sensitive to flucloxacillin. 100% of Streptococcus pneumoniae were sensitive to amoxicillin-clavulanic acid. While all S, pneumoniae, S. pyogenes and S. pyogenes were susceptible to penicillin and amoxicillin, overall only 36% of Gram positive organisms causing dacryocystitis were susceptible to penicillin or amoxicillin.

All Gram negative organisms identified were susceptible to gentamicin, TMP/SXT, piperacillin/tazobactam and ciprofloxacin. 50% of Gram negative organisms on which ceftriaxone was tested (Escherichia coli, Aggregatibacter species, Enterobacter cloacae, Hafnia alvei) were susceptible. Only Escherichia coli and E coliforms were universally susceptible to amoxicillin-clavulanic acid whilst Morganella morgagnii, Enterobacter cloacae and Hafnia alvei were all resistant.

The majority of patients [31/43, 72%] had complete resolution of dacryocystitis with no further episodes. Recurrence occurred in 8/43 [19%], and 4/43 [9%] were lost to follow up. Of those who experienced recurrence, 4 grew bacteria on their cultures, of which 2/4 [50%] were Staphylococcus aureus, 1/4 [25%] was Streptococcus pneumoniae and 1/4 [25%] was Pseudomonas aeruginosa. One patient had persisting symptoms with oral flucloxacillin therapy but was found to have MRSA and transitioned antibiotics to clindamycin which eventually resulted in resolution of dacryocystitis.

Acute dacryocystitis likely occurs due to bacterial overgrowth in the stagnant fluid of the lacrimal sac [4]. It can become a sight- or life-threatening infection with the potential to progress to orbital cellulitis, meningitis, and/or cavernous sinus thrombosis [10]. There are limited studies evaluating the microbiological profile of acute dacryocystitis, with the last study analysing dacryocystitis, both acute and chronic, in Australia being published in 2005 [8]. By providing an update on the microbiological landscape in dacryocystitis, targeted treatment can benefit the patient, particularly at a time of growing concerns regarding antimicrobial resistance [11].

The spectrum of bacterial pathogens and antibiotic susceptibility are reasonably consistent across various geographical regions[2]. Microbiological studies of dacryocystitis in Australia [8], North America [5], Iran [6, 7], China [1] and India [2, 4, 9] report that Staphylococcus aureus is the most common pathogen implicated in acute dacryocystitis, followed by Streptococcus species (including Streptococcus pneumoniae, Streptococcus pyogenes and Streptococcus anginosus). Razavi et al.’s earlier study from Iran identified Staphylococcus epidermidis as the most common cause of acute dacryocystitis [12], while Staphylococcus aureus is the predominant in more contemporary studies. S. epidermidis and other coagulase negative staphylococci are considered part of normal skin flora and therefore not routinely reported by our institution. Previous studies from Australia [8], China [1], India [4] and Iran [6] showed Pseudomonas aeruginosa and Haemophilus influenzae being common Gram negative isolates in dacryocystitis. In contrast, in this study, Escherichia coli and Aggregatibacter were the most common Gram negative organisms, with no cultures identifying Haemophilus influenzae. In this study, Gram positive isolates tended to predominate in acute dacryocystitis, which is consistent with much of the literature [2,3,4,5,6,7,8,9]. One of the unique cases in this series was a patient who had an acute dacryocystitis secondary to Hafnia alvei, a Gram negative enteric and oropharyngeal bacillus [13]. Infections caused by Hafnia alvei can occasionally be responsible for serious nosocomial and community acquired infections [14]. Cahill & Burns was the only paper in the literature that described finding more Gram negative isolates than Gram positive in patients with acute dacryocystitis, but this study was published in 1993 and the microbiology of dacryocystitis has likely changed since then [15]. Briscoe et al. found a predominance of Gram negative organisms, but the authors focused on purulent acute and chronic dacryocystitis, which likely represented a population with more virulent organisms [16].

MRSA infections are of increasing concern to clinical practitioners [17]. Risk factors include recent hospitalization, nursing home residence, use of injectable drugs, antibiotic usage, and chronic illness [18] should be screened appropriately. In this retrospective study, two patients had MRSA, representing 18% of Staphylococcus aureus infections causing dacryocystitis, which is similar to the incidence of MRSA for all ophthalmic infections (22%) reported from the same institution in 2012 [19]. MRSA has been found to be significantly associated with acute dacryocystitis as opposed to chronic presentations [5]. Dacryocystitis secondary to MRSA can be challenging to manage as it is more likely to be refractory to conservative treatment, and a definitive dacryocystorhinostomy (DCR) is often required for resolution [17]. In this study where there were two MRSA cases of acute dacryocystitis, one patient achieved resolution following an oral antibiotic course of clindamycin, while the other required an admission for IV vancomycin, a DCR within 1 week of admission, and was discharged onto oral trimethoprim-sulfamethoxazole. Both patients had resolution of dacryocystitis with no further recurrences documented.

In this study, the most common initial oral antibiotics prescribed were amoxicillin/clavulanic acid (13/20, 65%) and cefalexin (4/20, 20%). The current guidelines in Australia recommend cefalexin as the empirical treatment for dacryocystitis [20]. For the pathogens isolated in this study, 77% would have been susceptible to cefalexin (tested on 15 pathogens) and 82% would have been susceptible to flucloxacillin (tested on 11 pathogens). On the 17 pathogens in which TMP/SXT was tested, 100% were sensitive. Given the predominance of Gram positive organisms in dacryocystitis, antibiotics such as cephalosporins and amoxicillin/clavulanic acid may appear to be appropriate choices for initial antibiotic therapy, however, 34% of patients had their antibiotic treatment changed based on culture results. Furthermore, flucloxacillin alone may have too narrow a therapeutic range for empiric therapy as Gram negative organisms are not uncommonly identified in this condition. Flucloxacillin in combination with ceftriaxone is recommended for inpatients for coverage of periorbital cellulitis in the context of dacryocystitis, as is supported by local therapeutic guidelines [21]. This regimen would cover most of the organisms identified to cause dacryocystitis in this study, with the exception of the two cases of MRSA infection. Regarding Gram negative isolates in patients with purulent dacryocystitis, quinolones (e.g. ciprofloxacin), cephalosporins and aminoglycosides (e.g. gentamicin) have shown good coverage [4, 6]. If clinicians are concerned about a resistant organism, they may consider using amoxicillin/clavulanic acid as next-line empiric therapy [21]. TMP/SXT should also be considered if there is concern regarding MRSA or Gram negative organisms resistant to amoxicillin-clavulanic acid due to pathogens (e.g. Morganella morganii, Hafnia alvei, Klebsiella aerogenes, Proteus vulgaris) that constitutively produce amp C beta-lactamases [22].

The results obtained from the study was compared with the typical antibiotic resistance profile in antibiograms provided by our institution for common ophthalmic, otolaryngologic and upper respiratory pathogens identified in the same time period (Supplement 1). With the exception of TMP/SXT, the resistance profile of the organisms causing dacryocystitis in our study were consistent with those of the larger dataset representing typical antimicrobial susceptibilities in South Australia for similar samples.

There were several limitations of this study including its small sample size and retrospective nature. An additional limitation is the use of conjunctival swabs for specimen collection rather than lacrimal aspiration. Whilst there is a theoretical increase in risk of contamination when utilising conjunctival swabs [3], all attempts were made in this study to collect discharge from the region of the medial canthus and punctum as a representation of discharge secondary to a refluxable mucocele. Furthermore, performing a lacrimal aspirate can be a very painful procedure for the patient in the setting of an acute dacryocystitis. Nevertheless, it provides a snapshot of one tertiary Oculoplastic unit’s experience with acute dacryocystitis and the prevalence of microorganisms over a 5-year period. Another caveat in this study is that there was a notable proportion of cultures not being available for analysis, either due to culture-negative samples or no sample being taken (22/43, 51%). This was due to a number of factors, including commencement of empirical treatment in the community prior to presenting at our institution thus resulting in sterile aspirates, or no purulent discharge available to be swabbed.