Oral antibiotics for chronic blepharitis

Abstract

Background

Posterior blepharitis is common and causes ocular surface and lid damage as well as discomfort. It affects 37% to 47% of all ophthalmology patients; its incidence increasing with age. It is a multifactorial disease associated with multiple other pathologies, such as rosacea, meibomianitis, and infections. Treatment usually focuses on reliefing the symptoms by using artificial tears, lid scrubs, and warm compresses. The condition may be notoriously difficult to manage adequately once it becomes chronic. One such management approach for chronic blepharitis is the use of oral antibiotics for both their antibacterial as well as anti‐inflammatory properties. There are currently no guidelines regarding the use of oral antibiotics, including antibiotic type, dosage, and treatment duration, for the treatment of chronic blepharitis.

Objectives

To assess the benefits and harms of oral antibiotic use for people with chronic blepharitis.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (which contains the Cochrane Eyes and Vision Trials Register) (2020, Issue 8); Ovid MEDLINE; Embase.com; PubMed; Latin American and Caribbean Health Sciences Literature Database (LILACS); ClinicalTrials.gov, and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP). We did not use any date or language restrictions in the electronic search for trials. We last searched the electronic databases on 29 August 2020.

Selection criteria

We included randomized controlled trials (RCTs) comparing oral antibiotics with placebo in adult participants with chronic blepharitis (including staphylococcal, seborrhoeic, or Meibomian Gland Dysfunction (MGD)).

Data collection and analysis

We used standard Cochrane methodology and graded the certainty of the body of evidence for six outcomes using the GRADE classification.

Main results

We included two studies with 220 participants (numbers of eyes unclear). One parallel‐group RCT comparing oral doxycycline (40 mg once a day) with placebo enrolled 70 participants with blepharitis and facial rosacea in the USA. Follow‐up duration was three months. One three‐arm RCT conducted in South Korea investigated the effect of high‐dose (200 mg twice a day) and low‐dose (20 mg twice a day) doxycycline versus placebo after one month of study medication. It enrolled 50 participants with chronic MGD in each study arm (i.e. 150 participants enrolled in total).

The two studies did not evaluate the same outcome measurements, which precluded any meta‐analysis. The evidence for the effect of oral antibiotics on subjective improvement in symptoms was very uncertain. One study suggested that there was little to no effect of oral doxycycline on subjective symptoms based on the Ocular Surface Disease Index (OSDI) scores ranging from 0 to 100 (higher score indicates worse condition) (mean difference (MD) 3.55, 95% confidence interval (CI) −4.61 to 11.71; n = 70) and bulbar conjunctival hyperemia ranging from 0 (clear) to 4 (severe) (MD −0.01, 95% CI −0.38 to 0.36; n = 70) at 12 weeks. The three‐arm RCT showed that oral doxycycline may slightly improve number of symptoms (MD −0.56, 95% CI −0.95 to −0.17; n = 93 (high‐dose doxycycline versus placebo); MD −0.48, 95% CI −0.86 to −0.10; n = 93 (low‐dose doxycycline versus placebo)) and proportion of participants with symptom improvement (risk ratio (RR) 6.13, 95% CI 2.61 to 14.42; n = 93 (high‐dose doxycycline versus placebo); RR 6.54, 95% CI 2.79 to 15.30; n = 93 (low‐dose doxycycline versus placebo)) at one month, but the evidence is very uncertain. We judged the certainty of evidence for subjective symptoms as very low.

One study evaluated aqueous tear production by Schirmer's test (mm/5 min) (higher score indicates better condition) and tear film stability by measuring tear film break‐up time (TBUT) in seconds (higher score indicates better condition) at one month. We found very low certainty evidence that oral doxycycline may improve these clinical signs. The estimated MD in Schirmer's test score after one month of treatment was 4.09 mm (95% CI 2.38 to 5.80; n = 93) in the high‐dose doxycycline group versus the placebo group and 3.76 mm (95% CI 1.85 to 5.67; n = 93) in the low‐dose doxycycline group versus the placebo group. The estimated MD in TBUT after one month was 1.58 seconds (95% CI 0.57 to 2.59; n = 93) when comparing the high‐dose doxycycline group with the placebo group, and 1.70 seconds (95% CI 0.96 to 2.44; n = 93) when comparing the low‐dose doxycycline group with the placebo group. Although there was a noted improvement in these scores, their clinical importance remains uncertain.

One study suggested that oral doxycycline may increase the incidence of serious side effects: 18 (39%) participants in the high‐dose doxycycline group, 8 (17%) in the low‐dose doxycycline group, and 3 (6%) out of 47 participants in the placebo group experienced serious side effects (RR 6.13, 95% CI 1.94 to 19.41; n = 93 (high‐dose doxycycline versus placebo); RR 2.72, 95% CI 0.77 to 9.64; n = 93 (low‐dose doxycycline versus placebo)). Additionally, one study reported that one case of migraine headache and five cases of headache were observed in the oral doxycycline group, and one case of non‐Hodgkin's lymphoma was observed in the placebo group. We judged the certainty of evidence for adverse events as very low.

Authors' conclusions

There was insufficient evidence to draw any meaningful conclusions on the use of oral antibiotics for chronic blepharitis. Very low certainty evidence suggests that oral antibiotics may improve clinical signs, but may cause more adverse events. The evidence for the effect of oral antibiotics on subjective symptoms is very uncertain. Further trials are needed to provide high quality evidence on the use of oral antibiotics in the treatment of chronic blepharitis.

Plain language summary

Treating long‐lasting blepharitis (swollen and itchy eyelids): how well do antibiotics (medicines that kill bacteria) work when taken by mouth (orally)?

Key messages
· We did not find enough evidence to show how well antibiotics taken by mouth can treat long‐lasting blepharitis.

· One antibiotic tested may improve some clinical aspects (symptoms and course of the disease), but we are uncertain about its benefit in this respect, and it may also cause a higher number of unwanted effects.

· More studies are needed to find out how well antibiotics taken by mouth can treat long‐lasting blepharitis.

What is blepharitis?
Blepharitis is a common condition affecting the eyes. It causes swelling and redness on the edges of the eyelids, making them feel sore, and an itching or a gritty feeling in the eyes. The most common causes of blepharitis are infection by a type of bacteria that lives on the surface of the eye, or skin conditions such as dermatitis.

How is blepharitis treated?
Blepharitis is usually treated by regularly cleaning the eyelids, or using a cream or eye drops containing an antibiotic (a type of medicine that kills bacteria). If these do not work, then taking antibiotics by mouth (orally) is often tried. However, there are no guidelines about which type of antibiotic to give, what dose to use, or how long treatment should last.

What did we want to find out?
We wanted to find out how well antibiotics given by mouth can treat long‐lasting blepharitis.

What did we do?
We searched for studies that tested antibiotics given by mouth to treat long‐lasting blepharitis.

What did we find?
We found 2 studies in 220 adults with long‐lasting blepharitis. One study took place in the USA and lasted for three months; it tested the antibiotic doxycycline compared to placebo treatment (a 'dummy' treatment that does not contain any medicine but looks identical to the medicine being tested). The other study took place in South Korea and tested the effects of high or low doses of doxycycline compared to placebo treatment.

One study was funded by a pharmaceutical company; the other study did not report a source of funding. The studies measured results of treatment in different ways, so it was not possible to combine the studies' results to analyse them together.

What are the main results of our review?
We are uncertain about the effects of doxycycline on symptoms such as itching, burning, and watery eyes, as evaluated by people taking part in the studies.

One study measured changes in how much an affected eye watered (produced tears) before and after one month of treatment. High‐ and low‐dose doxycycline may improve the eye's ability to produce tears (evidence from 1 study with 93 people in each dose group).

One study evaluated dryness of the eye by measuring the time it took for a dry spot to appear on the surface of the eye after blinking (the 'tear film break‐up time'). Taking doxycycline (high‐ and low‐dose) for one month may improve problems with dryness (evidence from 1 study with 93 people in each dose group).

Taking doxycycline for one month may cause more unwanted effects than taking a placebo (evidence from 1 study in 139 people). The number of unwanted effects reported was higher in the high‐dose doxycycline group.

No studies measured:

· how many bacteria were present in the eye before or after treatment;

· people's well‐being (quality of life); or

· the costs and benefits of the treatments tested.

What are the limitations of the evidence?
We have very low confidence in the evidence because of limitations in the ways the studies were conducted, and because the results of the studies varied widely and were inconsistent. One study was funded by a pharmaceutical company, which could have affected the way the study was designed, conducted, and reported. Further research is likely to change and increase our confidence in the results.

How up‐to‐date is the evidence?
The evidence is current up to 29 August 2020.

Summary of findings

Summary of findings 1. Oral antibiotics compared with placebo for chronic blepharitis.

Oral antibiotics compared with placebo for chronic blepharitis
Patient or population: adult participants with chronic blepharitis Settings: tertiary care, university hospital, or private practice Intervention: oral antibiotics Comparison: placebo
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No. of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Placebo	Oral antibiotics
Subjective improvement in symptoms at 1 month and 3 months	OSDI, ranging from 0 to 100 (higher score indicates worse condition)		‐	70 (1 study)	⊕⊝⊝⊝ Very low1,2,3
	The mean  OSDI score decreased 8.70 at 3 months.	The mean decrease in OSDI score at 3 months was on average 3.55 smaller (95% CI −4.61 to 11.71).
	Bulbar conjunctival hyperemia, ranging from 0 (clear) to 4 (severe)		‐	70 (1 study)
	The mean  bulbar conjunctival hyperemia decreased 0.60 at 3 months.	The mean decrease in bulbar conjunctival hyperemia at 3 months was on average 0.01 greater (95% CI −0.38 to 0.36).
	Number of symptoms (higher score indicates worse condition)		‐	93 each analysis (1 study)
	The mean number of symptoms at 1 month was 2.01.	The mean number of symptoms at 1 month was on average 0.56 fewer (95% CI −0.95 to −0.17) in the high‐dose group and 0.48 fewer (95% CI −0.86 to −0.10) in the low‐dose group.
	Proportion of participants with improvement		RR 6.13 (95% CI 2.61 to 14.42) high dose vs placebo, RR 6.54 (95% CI 2.79 to 15.30) low dose vs placebo	93 each analysis (1 study)
	106 per 1000	652 per 1000 (95% CI 507 to 773) in the high‐dose group and 696 per 1000 (95% CI 552 to 809) in the low‐dose group
Measurements of eradication or decrease in numbers of colonies of positive cultures of bacteria	No data were available for this outcome.		‐	‐	‐
Clinical signs—Schirmer test scores at 1 month (mm/5 min)	Range: 0 to 20, higher scores indicate better conditions		‐	93 each analysis (1 study)	⊕⊝⊝⊝ Very low1,2
	The mean Schirmer test scores in the placebo group were 15.89.	The mean Schirmer test scores in the intervention group were on average 4.09 greater (95% CI 2.38 to 5.80) in the high‐dose group and 3.76 greater (95% CI 1.85 to 5.67) in the low‐dose group.
Clinical signs—TBUT at 1 month (seconds)	Range: 0 to 10, longer is better		‐	93 each analysis (1 study)	⊕⊝⊝⊝ Very low1,2
	The mean TBUT in the placebo group was 7.84	The mean TBUT in the intervention group was on average 1.58 greater (95% CI 0.57 to 2.59) in the high‐dose group and 1.70 greater (95% CI 0.96 to 2.44) in the low‐dose group.
Adverse events at 1 month	Cardinal side effects		RR 6.13 (95% CI 1.94 to 19.41) high dose vs placebo, RR 2.72 (95% CI 0.77 to 9.64) low dose vs placebo	93 each analysis (1 study)	⊕⊝⊝⊝ Very low1,2	Additional 1 study reported 1 case of migraine headache and 5 cases of headaches in the oral doxycycline group, and 1 case of non‐Hodgkin's lymphoma in the placebo group (NCT00560703).
	64 per 1000	391 per 1000 (high‐dose group) 174 per 1000 (low‐dose group)
Quality of life measures	No data were available for this outcome.		‐	‐	‐
Economic costs and benefits of different interventions	No data were available for this outcome.		‐	‐	‐
*The assumed risk is based on the estimate in the control group. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; OSDI: Ocular Surface Disease Index; RR: risk ratio; TBUT: tear film break‐up time
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

¹Downgraded one level for study limitations due to high risk of attrition or reporting bias among the included studies.
²Downgraded one or two levels for imprecision due to wide confidence intervals.
³Downgraded one level for inconsistency of results.

Background

Description of the condition

Blepharitis, defined as inflammation of the eyelids, is one of the most common ocular conditions and affects both children and adults (Lemp 2009; Viswalingham 2005). It can be classified either by anatomic area or by chronicity. Anterior blepharitis is defined as disease affecting the eyelid skin up to the lash base and follicles, while posterior blepharitis affects the meibomian glands (AAO 2018). Acute blepharitis, which responds relatively well to treatment, may be bacterial, viral, or parasitic in etiology (Eliason 2005), and is beyond the scope of this review. The more common form is chronic blepharitis, the clinical course of which tends to be indolent and recurring, distinguishing itself from the acute form. Chronic blepharitis can be further subdivided into three categories: staphylococcal, seborrhoeic, and meibomian gland dysfunction (MGD). However, the disease may also present as any combination of these three categories (AAO 2018). This review focuses on chronic blepharitis, and does not include blepharokeratoconjunctivitis, which has been evaluated by other Cochrane Reviews (O'Gallagher 2016; O'Gallagher 2017).

The two most common etiologies of anterior blepharitis are staphylococcal and seborrhoeic blepharitis. Staphylococcal blepharitis is believed to be associated with staphylococcal bacteria on the ocular surface, and is characterized by irritative redness of the lid margin that may have accompanying skin changes such as crusting or scaling. Toxins produced by certain strains of Staphylococcus aureus or Staphylococcus epidermidis may be a cause of irritation (Valenton 1973). Seborrheic blepharitis is characterized by less inflammation than staphylococcal blepharitis, but with more oily or greasy scaling. Some people with seborrhoeic blepharitis also exhibit characteristics of MGD.

Posterior blepharitis has various causes, such as MGD, infectious or allergic conjunctivitis, and systemic conditions such as acne rosacea (Nelson 2011). MGD is defined by the International Workshop on Meibomian Gland Dysfunction as "a chronic, diffuse abnormality of the meibomian glands, commonly characterized by terminal duct obstruction and/or qualitative/quantitative changes in the glandular secretion that may result in alteration of the tear film, symptoms of eye irritation, clinically apparent inflammation, and ocular surface disease" (Geerling 2011). Quantitative or qualitative deficiencies in meibum may be responsible for the symptoms experienced in MGD blepharitis. Hyperkeratinization of the meibomian gland epithelium (thickening of the lining of the glands) may lead to obstruction and a decrease in the quantity of meibomian gland secretions (Jester 1989a; Jester 1989b).

Demodex mites are also considered to be a causative factor for blepharitis (Czepita 2007). The mites, which infest the eyelid margin around the lash follicles and sebaceous glands, may have a role in both anterior and posterior blepharitis. Overgrowth of the mite population may be responsible for the ocular discomfort of blepharitis, as larger populations of the mites per cilia have been found to correlate with greater levels of ocular discomfort (Lee 2010).

Important comorbid conditions that predispose towards the development of blepharitis are acne rosacea, giant papillary conjunctivitis, and the use of the acne drug isotretinoin. Blepharitis may also result as a sequela of infections, various immunologic conditions, dermatologic conditions, toxic conditions, malignancy, and trauma (AAO 2018). These factors are numerous and beyond the scope of this review.

Epidemiology

Although not sight‐threatening, chronic blepharitis is one of the most common ocular disorders encountered by ophthalmologists (McCulley 2000). In a survey of US ophthalmologists and optometrists, 37% to 47% of people seen by those surveyed had signs of blepharitis (Lemp 2009). A 10‐year population‐based study of over one million people in South Korea found the incidence of clinically diagnosed blepharitis to be 1.1% per 100 person‐years, with an overall prevalence of 8.1%. The incidence and prevalence were higher in females and in those over 40 years old (Rim 2017). However, there have been too few epidemiologic studies to date to permit estimation of the true prevalence of blepharitis worldwide.

It has been postulated that staphylococcal blepharitis occurs more frequently in warmer climates (Bowman 1987). In a case‐control study conducted in the San Francisco Bay Area and Texas, staphylococcal blepharitis occurred more commonly in women, and had an average age of onset of 42 years (McCulley 1982; McCulley 1985). Approximately 25% to 50% of cases were associated with keratoconjunctivitis sicca (KCS), a class of dry eye syndrome (McCulley 1982; McCulley 1985).

In the same study, the mean age of participants with seborrhoeic blepharitis was 50 years (McCulley 1985). There was no difference in prevalence between men and women for this type of blepharitis. Ninety‐five per cent of participants with seborrhoeic blepharitis also had seborrhoeic dermatitis that presents with symptoms of flaking and greasy skin on the scalp, retroauricular area, glabella, and nasolabial folds (McCulley 1982); approximately one‐third of these participants had KCS (McCulley 1984).

The incidence of MGD increases with age (Driver 2005). The average age in the McCulley 1982 group of people with MGD blepharitis was 50 years old, and prevalence was equal between men and women (McCulley 1984). MGD may be more common in cooler climates (Bowman 1987). MGD seems to be more common in fair‐skinned individuals, but this may be due to its association with acne rosacea, which is also more prevalent in this population (Driver 1996). Acne rosacea is characterized by skin telangiectasias (dilated superficial blood vessels), erythema, papules, and pustules; it was diagnosed in 20% of people with MGD in the McCulley 1982 study. In the same study, 46% of people with MGD were diagnosed with seborrhoeic dermatitis (McCulley 1982).

There are two species of the Demodex mite that are known to be associated with human eye and skin diseases: Demodex folliculorum and Demodex brevis. D folliculorum primarily infests the infundibulum of hair follicles, while D brevis infests the deeper sebaceous structures and meibomian glands (Kabatas 2017). Demodex spp. are present in a quarter of the general population, and prevalence increases with age. Up to 100% of individuals over 70 years of age may have Demodex spp. present, leading to the assumption that it is commensal. However, there is a greater incidence of blepharitis in younger people who have the organism, and a higher number of mean organisms in people who have blepharitis (Biernat 2018). A meta‐analysis of studies in four countries showed a significant correlation between Demodex mite infestation and blepharitis (Zhao 2012). Blepharitis that has not responded to conventional treatments has been shown to improve with treatment of demodicosis. A recently published Cochrane Review evaluated the effects of tea tree oil on ocular Demodex infestation for people with Demodex blepharitis (Savla 2020). It concluded that the effectiveness of 5% to 50% tea tree oil is uncertain for the short‐term treatment of Demodex blepharitis.

In a study conducted in Florida in a cohort of people with any type of blepharitis, Groden 1991 found the prevalence of acne rosacea to be 44% and the prevalence of KCS to be 30%. In an Austrian study of 407 people with chronic blepharitis, 14.5% of participants had KCS, 32.9% had seborrhoeic dermatitis, and 26.7% had acne rosacea (Huber‐Spitzy 1991).

Presentation and diagnosis

General symptoms of blepharitis include burning, itchiness, gritty feeling of the eyes, contact lens intolerance, and photophobia (light‐sensitivity). Symptoms are usually worse in the morning, and a person with blepharitis may have several exacerbations and remissions.

Anterior blepharitis presents with redness and crusting of the eyelid margins. Specific examination of the eyelashes often reveals features such as poliosis (partial or total whitening of the lashes), madarosis (loss of eyelashes), trichiasis (misdirection of lashes towards the eye), and unusual features such as collarettes or dandruff‐like substances, which may all be helpful in diagnosing an etiologic pathogen. The eyelid skin or margin, or both, may be ulcerated or deformed in chronic cases.

Staphylococcal blepharitis is characterized on examination by erythema and edema of the eyelid margin. Telangiectasia may be present on the anterior eyelid. Brittle scales may be seen in the eyelashes, and these may form collarettes, which encircle the lash at the base or further up as the lash grows. In severe and long‐standing cases, trichiasis, poliosis, madarosis, eyelid ulceration, and eyelid and corneal scarring may occur (AAO 2018).

Seborrhoeic blepharitis is differentiated by less erythema, edema, and telangiectasia of the lid margins compared to staphylococcal blepharitis, but an increased amount of oily scales and greasy crusting on the lashes (McCulley 1985).

Posterior blepharitis may be seen clinically by examining the posterior eyelid margin. The meibomian glands may appear capped with oil, be dilated, or be visibly obstructed. Inflammation of the meibomian glands, termed 'meibomitis', may also be grossly observed during the examination in some cases, and may either be labelled as primary if with no other discernable cause, or secondary if deemed to be caused by nearby lesions such as lash base Demodex infestation. Meibomitis is an important clinical finding as if left untreated, it can cause inflammation of contacted ocular surface, leading to either an inflamed conjunctiva or cornea (meibomian keratoconjunctivitis). The secretions of the glands are usually turbid and thicker than normal. Telangiectasias and lid scarring may also be present in this area. In all forms of blepharitis, examination of the tear film may show instability and rapid evaporation.

Clinical exam findings in MGD include prominent palpebral conjunctival vessels and a frothy discharge along the lid margins. The meibomian gland orifices in people with MGD often present as pouted or plugged, and manual expression may result in a fluid that ranges from turbid to "cheeselike" in consistency (McCulley 1982).

The presence of cylindrical dandruff along the eyelash base is highly correlated to Demodex mite infestation, and this must always be a consideration in cases of blepharitis. Diagnosis of Demodex mite infestation requires epilation of two to three lashes, and visualization of the Demodex mite during direct observation of the lash base using a high‐powered microscope (Kabatas 2017; Lopez‐Ponce 2017).

Description of the intervention

Although the pathophysiology of anterior and posterior blepharitis may be different, the treatment options are similar. Current practice is such that people are generally offered treatment if they report discomfort, have visual symptoms, or have grossly visible lid lesions which may include scarring secondary to chronic disease. Children must be paid particular attention, as they often do not report chronic discomfort and may instead present with a history of chronic recurrent conjunctivitis and chalazion (small to large lumps on the eyelid, caused by obstruction of the secretory glands). In these cases, treatment is initiated in order to prevent chronic irritative damage to the eye, which may result in suboptimal vision in the long term. The American Academy of Ophthalmology (AAO) Preferred Practice Pattern guidelines emphasize the need for detection of blepharitis, as there is a significant overlap between blepharitis and other clinical entities, especially tear film dysfunction (AAO 2018). In such cases, a more customized treatment protocol may need to be developed to suit the person's unique requirements. Prior to treatment, it is important to inform the person that, although treatment may provide symptomatic relief, it may not necessarily cure the disease. A Cochrane Review of interventions for chronic blepharitis found that the available interventions did not show strong evidence for cure, and that long‐term monitoring may be necessary to provide optimal participant outcomes (Lindsley 2012).

Oral tetracycline or doxycycline may be effective for people with posterior blepharitis or symptoms not adequately controlled by lid hygiene and topical medications, especially those with concurrent MGD (AAO 2018). There are variations in clinical practice regarding the type and dosage of antibiotics for chronic blepharitis. Usual dosage can last from weeks to months. Clinical improvement requires several weeks, but once it is achieved, therapy may be discontinued or tapered to maintenance doses. The side effects of tetracycline may include raised intracranial hypertension, gastrointestinal disturbance, and skin hyperpigmentation. Furthermore, due to the affects on tooth and bone development, the drug has been assigned a Class D pregnancy category by the US Food and Drug Administration.

Improvement in posterior blepharitis with tetracyclines may be related to inhibition of bacterial lipases in both S aureus and S epidermidis (Dougherty 1991). Oral minocycline, a broad‐spectrum tetracycline antibiotic, showed some benefit in treating MGD in two case series (Aronowicz 2006; Shine 2003). However, it has shown side effects similar to other tetracyclines, such as photosensitivity, gastrointestinal disturbance, and hypersensitivity. In these cases, the AAO suggests the use of oral erythromycin as an alternative, particularly in children and pregnant women (AAO 2018). The use of oral azithromycin was investigated for its efficacy as a viable alternative (Zandian 2016). Tetracyclines such as doxycycline have the additional advantage of reducing dry eye symptoms by inducing tear production and reducing MGD, even though all the above‐mentioned drugs have similar effects in relieving subjective symptoms. A Cochrane Review on rosacea describes several drugs that may have potential use in the treatment of blepharitis, such as azelaic acid, brimonidine, and isotretinoin, but these have yet to be adequately investigated as treatments for blepharitis (van Zuuren 2015).

How the intervention might work

Current therapeutic options for chronic blepharitis aim to resolve symptoms, maintain comfort, and prevent complications that may result in suboptimal vision (AAO 2018; Pflugfelder 2014). In most people with chronic blepharitis, inflammation and dry eye disease is complicated by the existence of abnormal commensals, either in the form of bacteria or Demodex spp. mites. The eradication of these pathogenic factors aids the restoration of the normal lid and glandular microenvironment (AAO 2018; Liu 2010). Oral antibiotics help to eradicate the underlying infectious etiology, and may help to reduce inflammation.

Why it is important to do this review

Blepharitis is a common chronic disease whose etiology is poorly understood. The AAO Preferred Practice Pattern guidelines rate the strength of evidence to support oral tetracyclines for more severe and chronic cases as level III, a consensus opinion in the absence of substantial controlled evidence (AAO 2018). Additionally, since the literature search by the AAO encompassed only English‐language articles published between 1997 and 2017, a more complete systematic approach is warranted to identify trials and to highlight any evidence gaps in the literature. This review focused on the evidence to support oral antibiotics for chronic blepharitis.

Objectives

To assess the benefits and harms of oral antibiotic use for people with chronic blepharitis.

Methods

Criteria for considering studies for this review

Types of studies

We included only randomized controlled trials (RCTs).

Types of participants

We included trials in which participants were adults aged 16 years or older who were clinically diagnosed with chronic blepharitis, inclusive of staphylococcal, seborrhoeic, or MGD. We included adults only, as dosage for antibiotics varies between children and adults and did not provide an appropriately uniform comparator. As there are no standardized diagnostic protocols for chronic blepharitis or for the three subtypes, we also included studies where the type of chronic blepharitis was not specified, and studies that categorized chronic blepharitis using a different classification (e.g. meibomitis, primary meibomitis, meibomian keratoconjunctivitis). We included studies where participants with chronic blepharitis were a subset of the participants included in the study, if outcomes were reported separately for these participants. As many etiologies of chronic blepharitis have an inflammatory component, either locally (e.g. ocular adnexae) or systemically in dermatologic conditions (e.g. atopy or rosacea), many of these individuals are often treated with corticosteroid, either applied topically over the lids, or orally in the case of systemic involvement. We thus adjusted our search to include studies wherein additional treatment with antibiotics was compared to treatment without antibiotics in chronic blepharitis patients given corticosteroids for inflammation control.

Types of interventions

We included studies that made the following comparisons:

1. oral antibiotics versus placebo or no treatment;

2. oral antibiotics with corticosteroids versus corticosteroids alone.

Types of outcome measures

We included all relevant studies when we determined eligibility of studies for this review, irrespective of whether outcomes were reported.

Primary outcomes

Subjective improvement in symptoms, as judged by participant symptom report and measured by questionnaire, interview, or visual analogue scale (VAS), including but not limited to: irritation, burning, tearing, itching, eyelid sticking, photophobia, and increased frequency of blinking.

Since there were no standardized diagnostic criteria for blepharitis and no standardized scales on which to judge symptom severity, clinical improvement in symptoms was expected to vary among studies. Although studies should ideally use validated scales, all reported scales were considered for inclusion since standardized information was unavailable.

Secondary outcomes

1. Measurements of eradication or decrease in the numbers of colonies of positive cultures of bacteria

2. Clinical signs, including the following:

   1. aqueous tear production, measured by the mean change in Schirmer's test scores (mm/5 min) performed with or without topical ocular anesthesia;

   2. tear film stability, measured by the mean change in tear film break‐up time (TBUT) (seconds);

   3. ocular surface evaluation using physician's assessment and documentation by ocular surface staining score.

3. Adverse events, measured by participant report or changes in clinical findings

4. Quality of life measures

5. Economic costs and benefits of different interventions

Follow‐up time points

We planned that the time points for all outcomes would be one month, three months, and six months after treatment initiation. The ranges of follow‐up time points were from one month to three months in this review.

Search methods for identification of studies

Electronic searches

The Cochrane Eyes and Vision Information Specialist searched the following electronic databases for RCTs and controlled clinical trials. There were no language or publication year restrictions. We last searched the electronic databases on 29 August 2020.

• Cochrane Central Register of Controlled Trials (CENTRAL; 2020, Issue 8) (which contains the Cochrane Eyes and Vision Trials Register) in the Cochrane Library (searched 29 August 2020) (Appendix 1).

• MEDLINE Ovid (1946 to 29 August 2020) (Appendix 2).

• Embase.com (1947 to 29 August 2020) (Appendix 3).

• PubMed (1946 to 29 August 2020) (Appendix 4).

• LILACS (Latin American and Caribbean Health Science Information database) (1982 to 29 August 2020) (Appendix 5).

• US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov; searched 29 August 2020) (Appendix 6).

• World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp; searched 29 August 2020) (Appendix 7).

Searching other resources

We searched the reference lists of reports from trials we identified for additional trials. We contacted experts in the field for information on current, past, or unpublished trials. We did not conduct manual searches of conference proceedings or abstracts specifically for this review.

Data collection and analysis

We followed standard Cochrane methodology as described in our protocol.

Selection of studies

Two review authors (NO, SN) independently assessed the search results for potentially relevant trials based on our eligibility criteria (see Criteria for considering studies for this review), as defined in our published protocol. We obtained full‐text copies of all reports that were assessed as potentially eligible by at least one review author. Two review authors (NO, SN) independently reviewed the full‐text copies for inclusion in the review. We documented reports excluded at this stage, and noted the reasons for their exclusion. Any discrepancies were resolved by consensus.

Data extraction and management

Two review authors (NO, SN) independently extracted data from the reports of eligible trials onto data extraction forms. We extracted the following study characteristics: methods, participants, interventions, and outcomes. We included any relevant data, such as source of study funding and any conflicts of interest stated by the investigators, in the 'notes' category. We presented these study characteristics in table format. One review author (SN) entered the data into Review Manager 5 (Review Manager 2020), and a second review author (NO) verified the data entered. Any discrepancies were resolved by discussion. We extracted continuous and dichotomous data that were pertinent to the outcomes described in this protocol.

Assessment of risk of bias in included studies

Two review authors (NO, SN) independently assessed risk of bias in the included trials according to the methods in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2017). We based our assessment of studies on six domains:

1. selection bias (sequence generation and allocation concealment);

2. performance bias (masking of participants and researchers);

3. detection bias (masking of those responsible for assessing outcomes);

4. attrition bias (rates of follow‐up between groups and intention‐to‐treat (ITT) analysis);

5. reporting bias (selective outcome reporting).

For each risk of bias domain, two review authors (NO, SN) independently judged a given study to have a low, high, or unclear risk of bias. We presented descriptive documentation of the details of each parameter for each study in table format. We attempted to contact trial authors when additional data were necessary to evaluate bias, or when we judged the risk of bias to be unclear. However, we judged the parameter on the information available as we received no response to our request from the trial authors of one study, and we were unable to contact the trial authors for two studies.

Measures of treatment effect

We reported the mean difference (MD) and associated 95% confidence intervals (95% CI) for all continuous outcomes and rating scales. We planned to report the standardized mean difference (SMD) if the included studies reported the same outcomes but on different valid rating scales. We reported risk ratios (RR) and the relevant 95% CI for all dichotomous outcomes, including any rating scales that were dichotomized based on a clinically relevant cut‐off point.

Unit of analysis issues

The unit of analysis was the individual. Randomized cross‐over designs were inappropriate for this research question due to the risk of carry‐over effects and the possibility of period effects, and were thus excluded.

In future updates, if a study randomizes participants to a study intervention that is applied to both eyes, and performs the analysis separately by eye, we will classify the study as a cluster‐randomized study. For studies in which each eye of a participant receives a different study intervention, the unit of analysis should be the eye (intracomparative). In such situations, we will extract and include the data in our analyses only when the study has correctly taken the intraperson correlation into account in their analyses. If we include any multi‐arm studies, we will analyze them appropriately by splitting the control group into two or more groups.

Dealing with missing data

We attempted to contact study authors for additional information when data were missing or incomplete. We set the response time to two weeks; we used the data available in the published report as we either did not receive any reply in that time or were unable to contact study authors.

Assessment of heterogeneity

We did not assess the between‐study heterogeneity.

In future updates, we will assess clinical and methodological heterogeneity by examining variabilities in design, risk of bias, characteristics of participants, interventions, and outcomes among the included studies. We will perform the Chi² test and use the I² statistic to identify heterogeneity. We will interpret an I² greater than 50% as representing substantial heterogeneity.

Assessment of reporting biases

We planned to examine the symmetry of funnel plots to assess reporting biases if we could include 10 or more studies in a meta‐analysis. We did not examine the symmetry of funnel plots because we included two studies only.

Data synthesis

We qualitatively synthesized the current evidence based on the included studies.

In future updates, we will conduct meta‐analyses of studies with comparable outcomes and timing of outcomes if we judge the participants, interventions, and outcomes to be sufficiently similar. If there are fewer than three studies in a meta‐analysis and we do not detect any heterogeneity, we will use a fixed‐effect model. Otherwise, we will use a random‐effects model. We will document study results that are not compatible for meta‐analysis, and summarize the overall treatment effects as reported by each study.

Subgroup analysis and investigation of heterogeneity

We did not perform any subgroup analysis in this review. If sufficient data are provided in future updates, we will conduct subgroup analysis by analyzing anterior and posterior blepharitis separately, according to the classifications provided by the authors of the included studies.

Sensitivity analysis

We did not perform any sensitivity analysis. If sufficient data are available in future updates, we will conduct sensitivity analyses to examine the impact of excluding unpublished studies, industry‐funded studies, and studies with lower methodological quality.

Summary of findings and assessment of the certainty of the evidence

We prepared a summary of findings table comparing oral antibiotics versus placebo which presented the following six outcomes (Table 1).

1. Subjective improvement in symptoms

2. Measurements of eradication or decrease in numbers of colonies of positive cultures of bacteria

3. Clinical signs

4. Adverse events

5. Quality of life measures

6. Economic costs and benefits of different interventions

Two review authors (NO, SN) independently graded the certainty of the body of evidence for each outcome with a 4‐level rating scale (i.e. high, moderate, low, or very low) using the GRADE classification. Any disagreements were resolved by discussion and obtaining consensus within the review team. We downgraded the quality level of a body of evidence based on the following factors:

1. high risk of bias among included studies;

2. indirectness of evidence;

3. unexplained heterogeneity or inconsistency of results;

4. imprecision of results;

5. high probability of publication bias.

Results

Description of studies

A study flow diagram is shown in Figure 1.

1.

Study flow diagram.

Results of the search

Our search of the electronic databases on 29 August 2020 yielded 4382 records. After removing duplicates, we screened 3754 titles and abstracts. We retrieved 32 full‐text reports for further review. After full‐text screening, we excluded 28 studies (28 records) with reasons; identified two ongoing studies (two records); and included two studies (two records) in the review.

One of the ongoing studies started in 2015, and the findings have not been published yet (ChiCTR‐IPC‐15006715); the other ongoing study started in 2019, with an estimated completion date of December 2019 (NCT03953118).

Included studies

We included two studies with a total of 220 randomized participants. The numbers of eyes enrolled were unclear as the manner in which the study eye was chosen (e.g. average of two eyes, or one eye) was not reported. For details of the included studies, see Characteristics of included studies.

One parallel‐group RCT comparing oral doxycycline (40 mg once a day) with placebo enrolled 70 participants with blepharitis and facial rosacea in eight clinics in the USA (NCT00560703). Changes in bulbar conjunctival hyperemia and Ocular Surface Disease Index (OSDI) scores were reported after the study period of 12 weeks. This study was funded by industry.

One three‐arm RCT investigated the effect of high‐dose (200 mg twice a day) and low‐dose (20 mg twice a day) doxycycline versus placebo (Yoo 2005). It enrolled 50 participants with chronic meibomian gland dysfunction (MGD) in each arm (i.e. 150 participants in total) in South Korea. Subjective symptoms, Schirmer's test, and tear film break‐up time (TBUT) were measured after one month of study medication. This study failed to present sources of funding and conflicts of interest.

Excluded studies

We excluded 28 studies after full‐text review, and listed the reasons for exclusion in the Characteristics of excluded studies table. In summary, we excluded 14 studies based on study design (non‐RCTs or cross‐over trials), 11 studies on the interventions assessed (e.g. topical rather than oral antibiotics), and two studies due to ineligible study populations. We excluded one RCT that was terminated before enrollment.

Risk of bias in included studies

Risk of bias assessment for each included study is described in the Characteristics of included studies table. A summary of risk of bias assessments is shown in Figure 2.

2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Neither of the included studies described the method for allocation sequence generation and allocation concealment before assignment. We therefore judged the risk of selection bias to be unclear for both included studies.

Blinding

We judged both included studies as having a low risk of performance bias because in both studies participants and personnel were masked to intervention assignment. Outcome assessors were masked to the assignment in one study, which we assessed as at low risk of detection bias (NCT00560703). We judged the other included study as having an unclear risk of detection bias, as information on masking of outcome assessors was insufficient to permit a judgement of low or high risk of bias (Yoo 2005).

Incomplete outcome data

We judged one study in which there were no missing outcome data as at low risk of attrition bias (NCT00560703). We judged the other study as having a high risk of attrition bias because 11 (7.3%) of randomized participants were not included in the final analysis (Yoo 2005).

Selective reporting

Secondary outcomes that were originally presented in the trial registry were not presented in the final report in one study (NCT00560703), which we judged as at high risk of reporting bias. We judged the other study as at unclear risk of reporting bias because neither protocol nor trial registry was available for this study (Yoo 2005).

Effects of interventions

See: Table 1

We did not conduct any meta‐analysis or assess between‐study heterogeneity because the two included studies did not evaluate the same outcome measurements.

Primary outcomes

Subjective improvement in symptoms

NCT00560703 compared oral doxycycline (46 participants) with placebo (24 participants) and measured a subjective outcome based on the OSDI. Mean OSDI scores at 12 weeks postintervention decreased on average 5.15 (standard deviation (SD) 14.0) in the oral antibiotics group and 8.70 (SD 17.7) in the placebo group. The estimated mean difference (MD) between interventions was 3.55 (95% confidence interval (CI) −4.61 to 11.71; n = 70) (Analysis 1.1). Bulbar conjunctival hyperemia was also evaluated by using an ordinal variable ranging from 0 (clear) to 4 (severe). The mean change in bulbar conjunctival hyperemia after 12 weeks was −0.61 (SD 0.80) and −0.60 (SD 0.71) in the oral doxycycline and placebo groups, respectively. The estimated MD showed little to no difference between intervention groups (MD −0.01, 95% CI −0.38 to 0.36; n = 70) (Analysis 1.2).

1.1. Analysis.

Comparison 1: Oral antibiotics versus placebo, Outcome 1: Mean change in the Ocular Surface Disease Index at week 12

1.2. Analysis.

Comparison 1: Oral antibiotics versus placebo, Outcome 2: Mean change in bulbar conjunctival hyperemia at week 12

Yoo 2005 evaluated subjective symptoms by measuring the number of symptoms and proportion of participants with improvement. After one month of interventions, the mean number of symptoms decreased from 2.33 (SD 0.97) to 1.45 (SD 0.62) in the high‐dose doxycycline group (46 participants) and from 2.92 (SD 0.77) to 1.53 (SD 0.52) in the low‐dose doxycycline group (46 participants), compared with little to no change from 2.1 (SD 1.06) to 2.01 (1.23) in the placebo group (47 participants). The estimated MD in number of symptoms between groups at one month was −0.56 (95% CI −0.95 to −0.17; n = 93) between the high‐dose doxycycline and placebo groups, and MD −0.48 (95% CI −0.86 to −0.10; n = 93) between the low‐dose doxycycline and placebo groups, in favor of the oral doxycycline groups (Analysis 1.3). After one month, 30 (65.2%) of 46 participants in the high‐dose doxycycline group and 32 (69.6%) of 46 participants in the low‐dose doxycycline group experienced subjective improvement in symptoms compared with 5 (10.6%) of 47 participants in the placebo group. The estimated risk ratio (RR) for improvement in subjective symptoms was 6.13 (95% CI 2.61 to 14.42; n = 93) in the high‐dose doxycycline group compared to the placebo group, and 6.54 (95% CI 2.79 to 15.30; n = 93) in the low‐dose doxycycline group compared to the placebo group, both in favor of oral doxycycline (Analysis 1.4).

1.3. Analysis.

Comparison 1: Oral antibiotics versus placebo, Outcome 3: Mean number of symptoms at month 1

1.4. Analysis.

Comparison 1: Oral antibiotics versus placebo, Outcome 4: Subjective improvement in symptoms at month 1

We judged the certainty of the evidence for subjective symptoms as very low, downgrading one level for high risk of bias (−1), one level for imprecision due to wide confidence intervals (−1), and one level for inconsistency of results (−1).

Secondary outcomes

Measurements of eradication or decrease in numbers of colonies of positive cultures of bacteria

Neither of the included studies measured this outcome.

Clinical signs: Schirmer's test scores (mm/5 min)

One study evaluated the aqueous tear production by the Schirmer's test (Yoo 2005). Mean Schirmer's test scores (mm/5 minutes) increased to 19.98 mm (SD 4.05) after one month of treatment from 18.13 mm (SD 3.87) in the high‐dose doxycycline group (MD 1.85 mm, 95% CI 0.23 to 3.47; n = 46) and increased to 19.65 mm (SD 5.02) from 17.27 mm (SD 4.16) in the low‐dose doxycycline group (MD 2.38 mm, 95% CI 0.50 to 4.26; n = 46), whereas little to no improvement was observed in the control group (16.57 mm at baseline and 15.89 mm at one month; MD −0.68 mm, 95% CI −2.45 to 1.09; n = 47). The estimated MD in Schirmer's test score between groups after one month of treatment was 4.09 mm (95% CI 2.38 to 5.80; n = 93) in the high‐dose doxycycline group versus the placebo group and 3.76 mm (95% CI 1.85 to 5.67; n = 93) in the low‐dose doxycycline group versus the placebo group (Analysis 1.5).

1.5. Analysis.

Comparison 1: Oral antibiotics versus placebo, Outcome 5: Mean Schirmer's test score at month 1 (mm/5 min)

We judged the certainty of the evidence for this outcome as very low, downgrading one level for high risk of bias (−1) and two levels for imprecision due to wide confidence intervals (−2).

Clinical signs: tear film break‐up time (seconds)

Yoo and colleagues assessed the tear film stability by measuring TBUT (seconds) after one month of the study period (Yoo 2005). After one month, mean TBUT scores improved from 7.78 to 9.42 seconds (MD 1.64, 95% CI 0.62 to 2.66; n = 46) in the high‐dose doxycycline group, and from 7.82 to 9.54 seconds (MD 1.72, 95% CI 0.95 to 2.49; n = 46) in the low‐dose doxycycline group, although mean TBUT showed little to no improvement, from 7.80 to 7.84, in the placebo group (MD 0.04, 95% CI −0.80 to 0.88; n = 47). The estimated MD in TBUT between groups after one month was 1.58 seconds (95% CI 0.57 to 2.59; n = 93) when comparing the high‐dose doxycycline and placebo groups, and 1.70 seconds (95% CI 0.96 to 2.44; n = 93) when comparing the low‐dose doxycycline and placebo groups (Analysis 1.6). However, this study reported that no evidence of difference was found between low‐ and high‐dose doxycycline groups (P = 0.992; analysis of variance (ANOVA)).

1.6. Analysis.

Comparison 1: Oral antibiotics versus placebo, Outcome 6: Tear film break‐up time at month 1 (seconds) 

We judged the certainty of the evidence for this outcome as very low, downgrading one level for high risk of bias (−1) and two levels for imprecision due to wide confidence intervals (−2).

Clinical signs: ocular surface staining score

Neither of the included studies measured this outcome.

Adverse events

In NCT00560703, 1 (2.17%) participant experienced migraine headache (severe adverse event), and 5 (10.87%) participants reported headache in the oral doxycycline group, while 1 (4.17%) participant had non‐Hodgkin's lymphoma (severe adverse event) in the placebo group during the treatment period of 12 weeks.

In Yoo 2005, serious side effects were observed in 18 of 46 participants in the high‐dose doxycycline group, 8 of 46 participants in the low‐dose doxycycline group, and 3 of 47 participants in the placebo group (RR 6.13, 95% CI 1.94 to 19.41; n = 93 (high‐dose doxycycline versus placebo); RR 2.72, 95% CI 0.77 to 9.64; n = 93 (low‐dose doxycycline versus placebo)) (Analysis 1.7). The authors also reported that participants in the high‐dose doxycycline group tended to experience adverse events more frequently compared with those in low‐dose doxycycline group (P = 0.002; Chi² test). Four participants (8.7%) in the high‐dose doxycycline group, 2 (4.3%) in the low‐dose doxycycline group, and 1 (2.1%) in the placebo group withdrew from the study due to adverse events. Other common adverse events included gastrointestinal issues (e.g. dyspepsia, nausea, diarrhea) (21 participants), itchy skin, urticaria and erythematous papules (7 participants each), and stomatitis (1 participant).

1.7. Analysis.

Comparison 1: Oral antibiotics versus placebo, Outcome 7: Adverse events

We judged the certainty of the evidence for adverse events as very low, downgrading one level for high risk of bias (−1) and two levels for imprecision due to wide confidence intervals (−2).

Quality of life measures

Neither of the included studies evaluated quality of life measures.

Economic costs and benefits of different interventions

Neither of the included studies evaluated the costs of the interventions compared.

Discussion

Summary of main results

This systematic review summarized the data from two RCTs reporting on the efficacy of the use of oral antibiotics for chronic blepharitis. Both studies used tetracycline antibiotics (doxycycline) (NCT00560703; Yoo 2005). The studies comprised a total of 220 participants from two countries, the USA and South Korea.

The included studies examined different parameters in order to gauge effectiveness of treatment and, as such, a meta‐analysis was not performed.

The primary outcome for the two studies included subjective improvement of symptoms, although the outcome measurements varied betw• Summary of main results een studies. Use of oral antibiotics showed favorable effects on subjective symptoms in one study (Yoo 2005). The use of high‐ and low‐dose antibiotics demonstrated a slightly lower mean number of symptoms compared to placebo after a one‐month course of oral antibiotics. The other study examined subjective symptoms by measuring the Ocular Surface Disease Index (OSDI) and bulbar conjunctival hyperemia in 70 participants with blepharitis and facial rosacea (NCT00560703). The estimated mean difference showed little to no difference between the oral antibiotics group and the placebo group in OSDI scores and in bulbar conjunctival hyperemia after 12 weeks. The confidence interval was consistent with no difference. The lack of an important difference in ocular manifestations associated with facial rosacea highlights the importance of testing ophthalmologic effects of treatment specifically.

In Schirmer's testing, a mild improvement was noted in 46 participants treated with high‐dose doxycycline and in 46 participants treated with low‐dose doxycycline compared to the placebo control (Yoo 2005). This study also found a mild improvement in TBUT after one month in both high‐ and low‐dose doxycycline groups when compared to placebo. There was little to no difference between low‐ and high‐dose doxycycline groups. Although Schirmer's test and TBUT showed improvement in scores, the clinical importance of this improvement remains uncertain.

Adverse events were well documented in both studies. In NCT00560703, one severe adverse event (migraine headache) was recorded. The study by Yoo 2005 reported that adverse events increased in a dose‐dependent fashion. The authors reported that participants in high‐dose doxycycline group tended to experience adverse events more frequently compared with those in low‐dose doxycycline group. Eighteen of 46 participants in the high‐dose doxycycline group, eight of 46 participants in the low‐dose doxycycline group, and three of 47 participants in the placebo group experienced serious side effects. Four (8.7%) participants in the high‐dose doxycycline group, two (4.3%) participants in the low‐dose doxycycline group and one (2.1%) participant in the placebo group withdrew the study due to adverse events.

Neither of the included studies examined bacterial culture, ocular surface staining scores, quality of life measures, or economic cost and benefit.

The overall efficacy of oral antibiotics for chronic blepharitis showed mixed and uncertain results. Although beneficial effects may have been noted as outlined above, these effects were minimal and may not justify long‐term exposure to antibiotics and the corresponding risk of side effects.

The certainty of the evidence was very low due to high risk of bias and imprecision.

Overall completeness and applicability of evidence

The overall completeness and applicability of the evidence to address all of the objectives in this review are low. The included studies were geographically spread in the USA and Asia (South Korea), and the types of participants varied in conditions of chronic blepharitis (chronic blepharitis with facial rosacea and chronic meibomian gland dysfunction). However, although chronic blepharitis is a common ocular condition (affecting 37% to 47% of all ophthalmology patients seen based on a survey conducted in the USA) (Lemp 2009), we identified only two studies in this review, with relatively small sample sizes. With respect to interventions, only tetracycline was investigated in the included studies. In addition, the lack of standardized protocols for assessment of findings resulted in a wide variation of outcome measures between studies, hindering comparability.

Given these issues, the data gathered from this review may not be generalizable regarding the use of oral antibiotics for chronic blepharitis. 

Quality of the evidence

We graded the certainty of the evidence for all outcomes as very low due to high risk of bias and for imprecision due to wide confidence intervals.

There were limitations in study design and implementation in the included studies. In one study, 11 (7.3%) of randomized participants were excluded from the analysis, resulting in a judgement of high risk of attrition bias (Yoo 2005) In the other study, secondary outcomes which were originally described in the trial registry were not presented in the final report (NCT00560703). We judged this study as at high risk of reporting bias. Furthermore, each study had at least two risk of bias domains that were judged as unclear risk because information was insufficient to permit a judgement.

Potential biases in the review process

We followed standard Cochrane Review methods to minimize bias and Methodological Expectations of Cochrane Intervention Reviews (MECIR) standards for the reporting of new Cochrane Intervention Reviews (editorial-unit.cochrane.org/mecir) in conducting this review. An Information Specialist performed highly sensitive searches to identify studies. None of the review authors has any financial conflicts of interest.

Agreements and disagreements with other studies or reviews

The findings of this study are consistent with the American Academy of Ophthalmology's Preferred Practice Guidelines for blepharitis and published literature (AAO 2018; Geerling 2011; Nelson 2011; van Zuuren 2015). Based on very low certainty evidence, our analysis found mixed results with little to no difference between groups identified in one study, and a mild positive effect in the other study regarding symptom improvement with the use of oral antibiotics, yet a suggestive dose‐response relationship between oral doxycycline use and risks for adverse events.

The authors of a previous systematic review assessed the efficacy and safety of oral antibiotics in facial rosacea‐associated chronic blepharitis (van Zuuren 2015), and reported similarly inconclusive findings.

Authors' conclusions

Implications for practice.

The exact indications, dose requirement, and length of treatment for the use of oral antibiotics in the management of chronic blepharitis remains unclear. Based on existing clinical trial evidence, there may be some benefit but the evidence is very uncertain to permit a meaningful suggestion for this treatment approach. Safety and patient tolerability of using long‐term oral antibiotics, particularly tetracyclines, appears to be within acceptable margins, with encountered side effects being generally mild and consisting of general antibiotic‐associated side effects such as headache, nausea, and gastrointestinal distress. It is also important to note that systemic etiologies, such as facial rosacea, may necessitate the use of systemic antibiotics as part of their treatment. Other infectious etiologies, such as demodicosis, will carry their own etiology‐specific treatments as well. With respect to dose of oral antibiotics, the evidence does not clearly support a benefit to using high‐ versus low‐dose oral antibiotics for improvement in ophthalmologic signs and symptoms; on the other hand, the use of high‐dose antibiotic may result in a higher adverse effect profile. Consequently, monitoring patients remains important in order to screen for and respond early to any potential adverse event during treatment.

Implications for research.

Further research is required to provide high quality evidence on the use of oral antibiotics for the treatment of chronic blepharitis. Specifically, future studies would benefit from larger population sizes, standardized drug dosages, and the use of other antibiotics from different antibiotic classes, including macrolides (e.g. azithromycin, erythromycin, or clarithromycin), since tetracyclines are contraindicated in certain conditions such as pregnancy, to shed light on the overall efficacy of oral antibiotics for the treatment of chronic blepharitis. Scrutiny of etiology of chronic blepharitis is also crucial in future research, as different causes may respond differently to oral antibiotics due to the multifactorial nature of the disease. Some etiologies such as facial rosacea show good systemic response to antibiotic therapy, while others such as demodicosis require additional therapy to terminate lid infestations.

The development of a standardized method to measure ocular signs (e.g. tear film break‐up time and Schirmer's test) and the use of valid questionnaires to document subjective improvement in symptoms (e.g. eye discomfort) are also necessary to synthesize evidence across studies. Longitudinal assessment of intervention effects beyond one to three months may also provide valuable insights into treatment effects in the longer term. Furthermore, any measurement of clinical signs that inherently carry a high degree of subjectivity, such as eye redness, requires the development of rigidly worded descriptors, ideally with standardized reference photos. Lastly, future studies should be rigorously designed including sample sizes with appropriate calculation based on the estimated effects. Investigators should report adequate information by following the CONSORT statement for randomized controlled trials.

History

Protocol first published: Issue 8, 2020

Date	Event	Description
19 August 2008	Amended	Converted to new review format.

Appendices

Appendix 1. CENTRAL search strategy

#1 MeSH descriptor: [Blepharitis] explode all trees
#2 blephariti*
#3 demodicosis or demodicidosis or demodecosis or "cylindrical dandruff"
#4 demodex or "d. folliculorum" or "d. brevis"
#5 blepharoconjunctivitis
#6 ocular near (rosacea or mites)
#7 MeSH descriptor: [Meibomian Glands] explode all trees
#8 meibomian near gland*
#9 ocular near gland*
#10 (eye* or ocular or lid or lids) near inflamm*
#11 (eye* or ocular or lid or lids) near infect*
#12 (eye* or lid or lids) near seborrheic
#13 (eye* or lid or lids) near staphylococcal
#14 {OR #1‐#13}
#15 MeSH descriptor: [Anti‐Bacterial Agents] explode all trees
#16 antibiotic* OR (anti NEXT/1 biotic*) OR antibacterial* OR (anti NEXT/1 bacterial*)
#17 MeSH descriptor: [Erythromycin] explode all trees
#18 Erythromycin* OR abomacetin OR acneryne OR acnesol OR "akne mycin" OR "aknederm ery gel" OR aknemycin OR anamycin OR "bonac gel" OR "c solve 2" OR cliniderm OR deripil OR duraerythromycin OR "e mycin" OR "e base" OR "e glades" OR "e solve 2" OR emgel OR "emu v" OR "emu ve" OR emuvin OR emycin OR eriecu OR "erimycin t" OR eriprodin OR eritimix OR eritrex OR eritrocina OR eritromicina OR erixyl OR ermycin OR ermysin OR "ery maxin" OR "ery b" OR "ery diolan" OR "ery maxin" OR "ery tab" OR eryacne OR eryacnen OR eryc OR erycen OR erycette OR erycin OR erycinum OR eryderm OR erydermec OR erydermer OR eryfluid OR erygel OR eryhexal OR erymax OR erymaxin OR erymed OR erysafe OR erystrat OR erytab OR "eryth mycin" OR erythelan OR erythmycin OR erythomycin OR "erythra derm" OR erythran OR "erythro 200" OR "erythro teva" OR "erythro statin" OR erythrogan OR erythrogel OR erythrogran OR erythroguent OR erythromid OR erythromycine OR erythromycinum OR erythroteva OR erytop OR erytrarco OR erytrociclin OR etinycine OR etrolate OR etromycin OR ilocap OR ilocaps OR iloticina OR ilotycin OR "inderm gel" OR labocne OR latotryd OR lederpax OR mephamycin OR oftamolets OR pantodrin OR pantomycin OR pce OR pharyngocin OR primacine OR "r p mycin" OR robimycin OR romycin OR roymicin OR "rp mycin" OR rythocin OR "sans acne" OR sansac OR "skid gel e" OR staticin OR stiemycin OR stimycine OR "t stat" OR tstat OR theramycin OR "114‐07‐8" OR "70536‐18‐4"
#19 MeSH descriptor: [Azithromycin] explode all trees
#20 Azithromycin* OR aruzilina OR atizor OR azadose OR azasite OR azatril OR azenil OR azibiot OR azimin OR azithral OR azitrocin OR azitromax OR azitromicin OR azitromicina OR aziwok OR azomyne OR aztrin OR azydrop OR azyter OR azythromycin OR bazyt OR "cp 62933" OR "cp 62993" OR cp62933 OR cp62993 OR forcin OR goxal OR inedol OR infectoazit OR "isv 401" OR isv401 OR kromicin OR macrozit OR mezatrin OR octavax OR ordipha OR ribotrex OR sumamed OR sunamed OR tobyl OR toraseptol OR tromix OR trozocina OR ultreon OR vinzam OR xithrone OR "xz 450" OR xz450 OR zaret OR zarom OR zentavion OR zetamax OR zeto OR zibramax OR zifin OR zimericina OR zistic OR zithromax OR zithrox OR zitinn OR zitrim OR zitrobifan OR zitrocin OR zitromax OR zmax OR "117772‐70‐0" OR "121470‐24‐4" OR "83905‐01‐5"
#21 MeSH descriptor: [Macrolides] explode all trees
#22 Macrolide* OR macrotetrolide*
#23 MeSH descriptor: [Doxycycline] explode all trees
#24 Doxycyclin* OR atridox OR "BMY 28689" OR BMY28689 OR "BU 3839T" OR BU3839T OR "5 hydroxy 6 deoxytetracycline" OR "6 deoxy 5 hydroxytetracycline" OR "6 deoxy 5 oxytetracycline" OR "6 deoxyoxytetracycline" OR "6 desoxy 5 hydroxytetracycline" OR "6beta deoxyoxytetracycline" OR adoxa OR "alpha 6 desoxy 5 oxytetracycline" OR "alpha 6 deoxyoxytetracycline" OR amermycin OR atrax OR azudoxat OR bactidox OR banndoclin OR basedillin OR bassado OR biocolyn OR biodoxi OR bronmycin OR cloran OR cyclidox OR dentistar OR deoxycycline OR "deoxymycin dispersal" OR deoxymykoin OR deoxyoxytetracycline OR "desoxy oxytetracycline" OR desoxycycline OR doinmycin OR doryx OR dosil OR dotur OR doxaciclin OR doxacycline OR doxat OR doxatet OR "doxi‐sergo" OR doxibiotic OR doxicycline OR doxilin OR doximed OR doximycin OR doxin OR doxine OR doxocycline OR doxsig OR doxy OR doxybiocin OR doxycen OR doxychel OR doxycin OR "doxycydine monohydrate" OR doxylag OR doxylin OR doxymycin OR doxypuren OR doxytec OR doxytrim OR dumoxin OR duracycline OR esdoxin OR etidoxina OR gewacyclin OR "gs 3065" OR hydramycin OR ibralene OR idocyclin OR idocyklin OR interdoxin OR investin OR longamycin OR lydox OR magdrin OR medomycin OR mespafin OR mildox OR miraclin OR monodox OR nordox OR "nsc 56228" OR oracea OR paldomycin OR "pernox gel" OR periostat OR radox OR remycin OR respidox OR roximycin OR serodoxy OR servidoxine OR servidoxyne OR siadocin OR siclidon OR sigadoxin OR spanor OR supracyclin OR supramycina OR tenutan OR tolexine OR torymycin OR tsurupioxin OR unidox OR veemycin OR viadoxin OR "vibra s" OR "Vibra Tabs" OR vibrabiotic OR vibracina OR vibradox OR vibramicina OR vibramycin OR vibraveineuse OR vibravenos OR vibravet OR "viradoxyl‐n" OR wanmycin OR zadorin OR zenavod OR "10592‐13‐9" OR "17086‐28‐1" OR "564‐25‐0" OR "94088‐85‐4"
#25 MeSH descriptor: [Minocycline] explode all trees
#26 Minocyclin* OR "7 dimethylamino 6 demethyl 6 deoxytetracycline" OR akamin OR aknemin OR "akne puren" OR "aknin mino" OR aknosan OR arestin OR blemix OR borymycin OR cyclimycin OR cyclomin OR cyclops OR cynomycin OR dentomycin OR dynacin OR "Icht Oral" OR klinomycin OR klinotab OR lederderm OR logryx OR menocycline OR mestacine OR micromycin OR minakne OR minaxen OR "mino 50" OR "mino wolff" OR "minox 50" OR minocin OR minoclin OR minoclir OR minocyn OR minogalen OR minoline OR minolira OR minolis OR minomax OR minomycin OR minoplus OR minotab OR mirosin OR mynocine OR romin OR skinocyclin OR solodyn OR spicline OR vectran OR vectrin OR ximino OR "10118‐90‐8" OR "11006‐27‐2" OR "13614‐98‐7"
#27 MeSH descriptor: [Tetracyclines] explode all trees
#28 Tetracyclin*
#29 {OR #15‐#28}
#30 #14 AND #29 in Trials

Appendix 2. MEDLINE Ovid search strategy

1. Randomized Controlled Trial.pt.
2. Controlled Clinical Trial.pt.
3. (randomized or randomised).ab,ti.
4. placebo.ab,ti.
5. drug therapy.fs.
6. randomly.ab,ti.
7. trial.ab,ti.
8. groups.ab,ti.
9. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8
10. exp animals/ not humans.sh.
11. 9 not 10
12. exp BLEPHARITIS/
13. blephariti*.tw.
14. (demodicosis or demodicidosis or demodecosis or "cylindrical dandruff").tw.
15. (demodex or "d. folliculorum" or "d. brevis").tw.
16. blepharoconjunctivitis.tw.
17. (ocular adj3 (rosacea or mites)).tw.
18. exp Meibomian Glands/
19. (meibomian adj3 gland*).tw.
20. (ocular adj3 gland*).tw.
21. ((eye* or ocular or lid or lids) adj3 inflamm*).tw.
22. ((eye* or ocular or lid or lids) adj3 infect*).tw.
23. ((eye* or lid or lids) adj3 seborrheic).tw.
24. ((eye* or lid or lids) adj3 staphylococcal).tw.
25. or/12‐24
26. exp Anti‐Bacterial Agents/
27. (antibiotic* or (anti adj1 biotic*) or antibacterial* or (anti adj1 bacterial*)).tw.
28. exp Erythromycin/
29. (Erythromycin* or abomacetin or acneryne or acnesol or "akne mycin" or "aknederm ery gel" or aknemycin or anamycin or "bonac gel" or "c solve 2" or cliniderm or deripil or duraerythromycin or "e mycin" or "e base" or "e glades" or "e solve 2" or emgel or "emu v" or "emu ve" or emuvin or emycin or eriecu or "erimycin t" or eriprodin or eritimix or eritrex or eritrocina or eritromicina or erixyl or ermycin or ermysin or "ery maxin" or "ery b" or "ery diolan" or "ery maxin" or "ery tab" or eryacne or eryacnen or eryc or erycen or erycette or erycin or erycinum or eryderm or erydermec or erydermer or eryfluid or erygel or eryhexal or erymax or erymaxin or erymed or erysafe or erystrat or erytab or "eryth mycin" or erythelan or erythmycin or erythomycin or "erythra derm" or erythran or "erythro 200" or "erythro teva" or "erythro statin" or erythrogan or erythrogel or erythrogran or erythroguent or erythromid or erythromycine or erythromycinum or erythroteva or erytop or erytrarco or erytrociclin or etinycine or etrolate or etromycin or ilocap or ilocaps or iloticina or ilotycin or "inderm gel" or labocne or latotryd or lederpax or mephamycin or oftamolets or pantodrin or pantomycin or pce or pharyngocin or primacine or "r p mycin" or robimycin or romycin or roymicin or "rp mycin" or rythocin or "sans acne" or sansac or "skid gel e" or staticin or stiemycin or stimycine or "t stat" or tstat or theramycin or "114‐07‐8" or "70536‐18‐4").tw,rn.
30. exp Azithromycin/
31. (Azithromycin* or aruzilina or atizor or azadose or azasite or azatril or azenil or azibiot or azimin or azithral or azitrocin or azitromax or azitromicin or azitromicina or aziwok or azomyne or aztrin or azydrop or azyter or azythromycin or bazyt or "cp 62933" or "cp 62993" or cp62933 or cp62993 or forcin or goxal or inedol or infectoazit or "isv 401" or isv401 or kromicin or macrozit or mezatrin or octavax or ordipha or ribotrex or sumamed or sunamed or tobyl or toraseptol or tromix or trozocina or ultreon or vinzam or xithrone or "xz 450" or xz450 or zaret or zarom or zentavion or zetamax or zeto or zibramax or zifin or zimericina or zistic or zithromax or zithrox or zitinn or zitrim or zitrobifan or zitrocin or zitromax or zmax or "117772‐70‐0" or "121470‐24‐4" or "83905‐01‐5").tw,rn.
32. exp Macrolides/
33. (Macrolide* or macrotetrolide*).tw.
34. exp Doxycycline/
35. (Doxycyclin* or atridox or "BMY 28689" or BMY28689 or "BU 3839T" or BU3839T or "5 hydroxy 6 deoxytetracycline" or "6 deoxy 5 hydroxytetracycline" or "6 deoxy 5 oxytetracycline" or "6 deoxyoxytetracycline" or "6 desoxy 5 hydroxytetracycline" or "6beta deoxyoxytetracycline" or adoxa or "alpha 6 desoxy 5 oxytetracycline" or "alpha 6 deoxyoxytetracycline" or amermycin or atrax or azudoxat or bactidox or banndoclin or basedillin or bassado or biocolyn or biodoxi or bronmycin or cloran or cyclidox or dentistar or deoxycycline or "deoxymycin dispersal" or deoxymykoin or deoxyoxytetracycline or "desoxy oxytetracycline" or desoxycycline or doinmycin or doryx or dosil or dotur or doxaciclin or doxacycline or doxat or doxatet or "doxi‐sergo" or doxibiotic or doxicycline or doxilin or doximed or doximycin or doxin or doxine or doxocycline or doxsig or doxy or doxybiocin or doxycen or doxychel or doxycin or "doxycydine monohydrate" or doxylag or doxylin or doxymycin or doxypuren or doxytec or doxytrim or dumoxin or duracycline or esdoxin or etidoxina or gewacyclin or "gs 3065" or hydramycin or ibralene or idocyclin or idocyklin or interdoxin or investin or longamycin or lydox or magdrin or medomycin or mespafin or mildox or miraclin or monodox or nordox or "nsc 56228" or oracea or paldomycin or "pernox gel" or periostat or radox or remycin or respidox or roximycin or serodoxy or servidoxine or servidoxyne or siadocin or siclidon or sigadoxin or spanor or supracyclin or supramycina or tenutan or tolexine or torymycin or tsurupioxin or unidox or veemycin or viadoxin or "vibra s" or "Vibra Tabs" or vibrabiotic or vibracina or vibradox or vibramicina or vibramycin or vibraveineuse or vibravenos or vibravet or "viradoxyl‐n" or wanmycin or zadorin or zenavod or "10592‐13‐9" or "17086‐28‐1" or "564‐25‐0" or "94088‐85‐4").tw,rn.
36. exp Minocycline/
37. (Minocyclin* or "7 dimethylamino 6 demethyl 6 deoxytetracycline" or akamin or aknemin or "akne puren" or "aknin mino" or aknosan or arestin or blemix or borymycin or cyclimycin or cyclomin or cyclops or cynomycin or dentomycin or dynacin or "Icht Oral" or klinomycin or klinotab or lederderm or logryx or menocycline or mestacine or micromycin or minakne or minaxen or "mino 50" or "mino wolff" or "minox 50" or minocin or minoclin or minoclir or minocyn or minogalen or minoline or minolira or minolis or minomax or minomycin or minoplus or minotab or mirosin or mynocine or romin or skinocyclin or solodyn or spicline or vectran or vectrin or ximino or "10118‐90‐8" or "11006‐27‐2" or "13614‐98‐7").tw,rn.
38. exp Tetracyclines/
39. Tetracyclin*.tw.
40. or/26‐39
41. 25 and 40
42. 11 and 41

The search filter for trials at the beginning of the MEDLINE strategy is from the published paper by Glanville 2006 .

Appendix 3. Embase.com search strategy

#1 'randomized controlled trial'/exp
#2 'randomization'/exp
#3 'double blind procedure'/exp
#4 'single blind procedure'/exp
#5 random*:ab,ti
#6 #1 OR #2 OR #3 OR #4 OR #5
#7 'animal'/exp OR 'animal experiment'/exp
#8 'human'/exp
#9 #7 AND #8
#10 #7 NOT #9
#11 #6 NOT #10
#12 'clinical trial'/exp
#13 (clin* NEAR/3 trial*):ab,ti
#14 ((singl* OR doubl* OR trebl* OR tripl*) NEAR/3 (blind* OR mask*)):ab,ti
#15 'placebo'/exp
#16 placebo*:ab,ti
#17 random*:ab,ti
#18 'experimental design'/exp
#19 'crossover procedure'/exp
#20 'control group'/exp
#21 'latin square design'/exp
#22 #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21
#23 #22 NOT #10
#24 #23 NOT #11
#25 'comparative study'/exp
#26 'evaluation'/exp
#27 'prospective study'/exp
#28 control*:ab,ti OR prospectiv*:ab,ti OR volunteer*:ab,ti
#29 #25 OR #26 OR #27 OR #28
#30 #29 NOT #10
#31 #30 NOT (#11 OR #23)
#32 #11 OR #24 OR #31
#33 'blepharitis'/exp
#34 blephariti*:ti,ab,kw
#35 demodicosis:ti,ab,kw OR demodicidosis:ti,ab,kw OR demodecosis:ti,ab,kw OR 'cylindrical dandruff':ti,ab,kw
#36 demodex:ti,ab,kw OR 'd. folliculorum':ti,ab,kw OR 'd. brevis':ti,ab,kw
#37 blepharoconjunctivitis.:ti,ab,kw
#38 (ocular NEAR/3 (rosacea OR mites)):ab,ti,kw
#39 'meibomian gland'/exp
#40 (meibomian NEAR/3 gland*):ab,ti,kw
#41 (ocular NEAR/3 gland*):ti,ab,kw
#42 ((eye* OR ocular OR lid OR lids) NEAR/3 inflamm*):ti,ab,kw
#43 ((eye* OR ocular OR lid OR lids) NEAR/3 infect*):ti,ab,kw
#44 ((eye* OR lid OR lids) NEAR/3 seborrheic):ti,ab,kw
#45 ((eye* OR lid OR lids) NEAR/3 staphylococcal):ti,ab,kw
#46 #33 OR #34 OR #35 OR #36 OR #37 OR #38 OR #39 OR #40 OR #41 OR #42 OR #43 OR #44 OR #45
#47 'antibiotic agent'/de
#48 antibiotic*:ab,ti,kw OR ((anti NEXT/1 biotic*):ab,ti,kw) OR antibacterial*:ab,ti,kw OR ((anti NEXT/1 bacterial*):ab,ti,kw)
#49 'erythromycin'/exp
#50 erythromycin*:ab,ti,kw,tn OR abomacetin:ab,ti,kw,tn OR acneryne:ab,ti,kw,tn OR acnesol:ab,ti,kw,tn OR 'akne mycin':ab,ti,kw,tn OR 'aknederm ery gel':ab,ti,kw,tn OR aknemycin:ab,ti,kw,tn OR anamycin:ab,ti,kw,tn OR 'bonac gel':ab,ti,kw,tn OR 'c solve 2':ab,ti,kw,tn OR cliniderm:ab,ti,kw,tn OR deripil:ab,ti,kw,tn OR duraerythromycin:ab,ti,kw,tn OR 'e mycin':ab,ti,kw,tn OR 'e base':ab,ti,kw,tn OR 'e glades':ab,ti,kw,tn OR 'e solve 2':ab,ti,kw,tn OR emgel:ab,ti,kw,tn OR 'emu v':ab,ti,kw,tn OR 'emu ve':ab,ti,kw,tn OR emuvin:ab,ti,kw,tn OR emycin:ab,ti,kw,tn OR eriecu:ab,ti,kw,tn OR 'erimycin t':ab,ti,kw,tn OR eriprodin:ab,ti,kw,tn OR eritimix:ab,ti,kw,tn OR eritrex:ab,ti,kw,tn OR eritrocina:ab,ti,kw,tn OR eritromicina:ab,ti,kw,tn OR erixyl:ab,ti,kw,tn OR ermycin:ab,ti,kw,tn OR ermysin:ab,ti,kw,tn OR 'ery b':ab,ti,kw,tn OR 'ery diolan':ab,ti,kw,tn OR 'ery maxin':ab,ti,kw,tn OR 'ery tab':ab,ti,kw,tn OR eryacne:ab,ti,kw,tn OR eryacnen:ab,ti,kw,tn OR eryc:ab,ti,kw,tn OR erycen:ab,ti,kw,tn OR erycette:ab,ti,kw,tn OR erycin:ab,ti,kw,tn OR erycinum:ab,ti,kw,tn OR eryderm:ab,ti,kw,tn OR erydermec:ab,ti,kw,tn OR erydermer:ab,ti,kw,tn OR eryfluid:ab,ti,kw,tn OR erygel:ab,ti,kw,tn OR eryhexal:ab,ti,kw,tn OR erymax:ab,ti,kw,tn OR erymaxin:ab,ti,kw,tn OR erymed:ab,ti,kw,tn OR erysafe:ab,ti,kw,tn OR erystrat:ab,ti,kw,tn OR erytab:ab,ti,kw,tn OR 'eryth mycin':ab,ti,kw,tn OR erythelan:ab,ti,kw,tn OR erythmycin:ab,ti,kw,tn OR erythomycin:ab,ti,kw,tn OR 'erythra derm':ab,ti,kw,tn OR erythran:ab,ti,kw,tn OR 'erythro 200':ab,ti,kw,tn OR 'erythro teva':ab,ti,kw,tn OR 'erythro statin':ab,ti,kw,tn OR erythrogan:ab,ti,kw,tn OR erythrogel:ab,ti,kw,tn OR erythrogran:ab,ti,kw,tn OR erythroguent:ab,ti,kw,tn OR erythromid:ab,ti,kw,tn OR erythromycine:ab,ti,kw,tn OR erythromycinum:ab,ti,kw,tn OR erythroteva:ab,ti,kw,tn OR erytop:ab,ti,kw,tn OR erytrarco:ab,ti,kw,tn OR erytrociclin:ab,ti,kw,tn OR etinycine:ab,ti,kw,tn OR etrolate:ab,ti,kw,tn OR etromycin:ab,ti,kw,tn OR ilocap:ab,ti,kw,tn OR ilocaps:ab,ti,kw,tn OR iloticina:ab,ti,kw,tn OR ilotycin:ab,ti,kw,tn OR 'inderm gel':ab,ti,kw,tn OR labocne:ab,ti,kw,tn OR latotryd:ab,ti,kw,tn OR lederpax:ab,ti,kw,tn OR mephamycin:ab,ti,kw,tn OR oftamolets:ab,ti,kw,tn OR pantodrin:ab,ti,kw,tn OR pantomycin:ab,ti,kw,tn OR pce:ab,ti,kw,tn OR pharyngocin:ab,ti,kw,tn OR primacine:ab,ti,kw,tn OR 'r p mycin':ab,ti,kw,tn OR robimycin:ab,ti,kw,tn OR romycin:ab,ti,kw,tn OR roymicin:ab,ti,kw,tn OR 'rp mycin':ab,ti,kw,tn OR rythocin:ab,ti,kw,tn OR 'sans acne':ab,ti,kw,tn OR sansac:ab,ti,kw,tn OR 'skid gel e':ab,ti,kw,tn OR staticin:ab,ti,kw,tn OR stiemycin:ab,ti,kw,tn OR stimycine:ab,ti,kw,tn OR 't stat':ab,ti,kw,tn OR tstat:ab,ti,kw,tn OR theramycin:ab,ti,kw,tn OR '114‐07‐8':ab,ti,kw,tn OR '70536‐18‐4':ab,ti,kw,tn
#51 'azithromycin'/exp
#52 azithromycin*:ab,ti,kw,tn OR aruzilina:ab,ti,kw,tn OR atizor:ab,ti,kw,tn OR azadose:ab,ti,kw,tn OR azasite:ab,ti,kw,tn OR azatril:ab,ti,kw,tn OR azenil:ab,ti,kw,tn OR azibiot:ab,ti,kw,tn OR azimin:ab,ti,kw,tn OR azithral:ab,ti,kw,tn OR azitrocin:ab,ti,kw,tn OR azitromax:ab,ti,kw,tn OR azitromicin:ab,ti,kw,tn OR azitromicina:ab,ti,kw,tn OR aziwok:ab,ti,kw,tn OR azomyne:ab,ti,kw,tn OR aztrin:ab,ti,kw,tn OR azydrop:ab,ti,kw,tn OR azyter:ab,ti,kw,tn OR azythromycin:ab,ti,kw,tn OR bazyt:ab,ti,kw,tn OR 'cp 62933':ab,ti,kw,tn OR 'cp 62993':ab,ti,kw,tn OR cp62933:ab,ti,kw,tn OR cp62993:ab,ti,kw,tn OR forcin:ab,ti,kw,tn OR goxal:ab,ti,kw,tn OR inedol:ab,ti,kw,tn OR infectoazit:ab,ti,kw,tn OR 'isv 401':ab,ti,kw,tn OR isv401:ab,ti,kw,tn OR kromicin:ab,ti,kw,tn OR macrozit:ab,ti,kw,tn OR mezatrin:ab,ti,kw,tn OR octavax:ab,ti,kw,tn OR ordipha:ab,ti,kw,tn OR ribotrex:ab,ti,kw,tn OR sumamed:ab,ti,kw,tn OR sunamed:ab,ti,kw,tn OR tobyl:ab,ti,kw,tn OR toraseptol:ab,ti,kw,tn OR tromix:ab,ti,kw,tn OR trozocina:ab,ti,kw,tn OR ultreon:ab,ti,kw,tn OR vinzam:ab,ti,kw,tn OR xithrone:ab,ti,kw,tn OR 'xz 450':ab,ti,kw,tn OR xz450:ab,ti,kw,tn OR zaret:ab,ti,kw,tn OR zarom:ab,ti,kw,tn OR zentavion:ab,ti,kw,tn OR zetamax:ab,ti,kw,tn OR zeto:ab,ti,kw,tn OR zibramax:ab,ti,kw,tn OR zifin:ab,ti,kw,tn OR zimericina:ab,ti,kw,tn OR zistic:ab,ti,kw,tn OR zithromax:ab,ti,kw,tn OR zithrox:ab,ti,kw,tn OR zitinn:ab,ti,kw,tn OR zitrim:ab,ti,kw,tn OR zitrobifan:ab,ti,kw,tn OR zitrocin:ab,ti,kw,tn OR zitromax:ab,ti,kw,tn OR zmax:ab,ti,kw,tn OR '117772‐70‐0':ab,ti,kw,tn OR '121470‐24‐4':ab,ti,kw,tn OR '83905‐01‐5':ab,ti,kw,tn
#53 'macrolide'/exp
#54 Macrolide*:ab,ti,kw,tn OR macrotetrolide*:ab,ti,kw,tn
#55 'doxycycline'/exp
#56 doxycyclin*:ab,ti,kw,tn OR atridox:ab,ti,kw,tn OR 'bmy 28689':ab,ti,kw,tn OR bmy28689:ab,ti,kw,tn OR 'bu 3839t':ab,ti,kw,tn OR bu3839t:ab,ti,kw,tn OR '5 hydroxy 6 deoxytetracycline':ab,ti,kw,tn OR '6 deoxy 5 hydroxytetracycline':ab,ti,kw,tn OR '6 deoxy 5 oxytetracycline':ab,ti,kw,tn OR '6 deoxyoxytetracycline':ab,ti,kw,tn OR '6 desoxy 5 hydroxytetracycline':ab,ti,kw,tn OR '6beta deoxyoxytetracycline':ab,ti,kw,tn OR adoxa:ab,ti,kw,tn OR 'alpha 6 desoxy 5 oxytetracycline':ab,ti,kw,tn OR 'alpha 6 deoxyoxytetracycline':ab,ti,kw,tn OR amermycin:ab,ti,kw,tn OR atrax:ab,ti,kw,tn OR azudoxat:ab,ti,kw,tn OR bactidox:ab,ti,kw,tn OR banndoclin:ab,ti,kw,tn OR basedillin:ab,ti,kw,tn OR bassado:ab,ti,kw,tn OR biocolyn:ab,ti,kw,tn OR biodoxi:ab,ti,kw,tn OR bronmycin:ab,ti,kw,tn OR cloran:ab,ti,kw,tn OR cyclidox:ab,ti,kw,tn OR dentistar:ab,ti,kw,tn OR deoxycycline:ab,ti,kw,tn OR 'deoxymycin dispersal':ab,ti,kw,tn OR deoxymykoin:ab,ti,kw,tn OR deoxyoxytetracycline:ab,ti,kw,tn OR 'desoxy oxytetracycline':ab,ti,kw,tn OR desoxycycline:ab,ti,kw,tn OR doinmycin:ab,ti,kw,tn OR doryx:ab,ti,kw,tn OR dosil:ab,ti,kw,tn OR dotur:ab,ti,kw,tn OR doxaciclin:ab,ti,kw,tn OR doxacycline:ab,ti,kw,tn OR doxat:ab,ti,kw,tn OR doxatet:ab,ti,kw,tn OR 'doxi‐sergo':ab,ti,kw,tn OR doxibiotic:ab,ti,kw,tn OR doxicycline:ab,ti,kw,tn OR doxilin:ab,ti,kw,tn OR doximed:ab,ti,kw,tn OR doximycin:ab,ti,kw,tn OR doxin:ab,ti,kw,tn OR doxine:ab,ti,kw,tn OR doxocycline:ab,ti,kw,tn OR doxsig:ab,ti,kw,tn OR doxy:ab,ti,kw,tn OR doxybiocin:ab,ti,kw,tn OR doxycen:ab,ti,kw,tn OR doxychel:ab,ti,kw,tn OR doxycin:ab,ti,kw,tn OR 'doxycydine monohydrate':ab,ti,kw,tn OR doxylag:ab,ti,kw,tn OR doxylin:ab,ti,kw,tn OR doxymycin:ab,ti,kw,tn OR doxypuren:ab,ti,kw,tn OR doxytec:ab,ti,kw,tn OR doxytrim:ab,ti,kw,tn OR dumoxin:ab,ti,kw,tn OR duracycline:ab,ti,kw,tn OR esdoxin:ab,ti,kw,tn OR etidoxina:ab,ti,kw,tn OR gewacyclin:ab,ti,kw,tn OR 'gs 3065':ab,ti,kw,tn OR hydramycin:ab,ti,kw,tn OR ibralene:ab,ti,kw,tn OR idocyclin:ab,ti,kw,tn OR idocyklin:ab,ti,kw,tn OR interdoxin:ab,ti,kw,tn OR investin:ab,ti,kw,tn OR longamycin:ab,ti,kw,tn OR lydox:ab,ti,kw,tn OR magdrin:ab,ti,kw,tn OR medomycin:ab,ti,kw,tn OR mespafin:ab,ti,kw,tn OR mildox:ab,ti,kw,tn OR miraclin:ab,ti,kw,tn OR monodox:ab,ti,kw,tn OR nordox:ab,ti,kw,tn OR 'nsc 56228':ab,ti,kw,tn OR oracea:ab,ti,kw,tn OR paldomycin:ab,ti,kw,tn OR 'pernox gel':ab,ti,kw,tn OR periostat:ab,ti,kw,tn OR radox:ab,ti,kw,tn OR remycin:ab,ti,kw,tn OR respidox:ab,ti,kw,tn OR roximycin:ab,ti,kw,tn OR serodoxy:ab,ti,kw,tn OR servidoxine:ab,ti,kw,tn OR servidoxyne:ab,ti,kw,tn OR siadocin:ab,ti,kw,tn OR siclidon:ab,ti,kw,tn OR sigadoxin:ab,ti,kw,tn OR spanor:ab,ti,kw,tn OR supracyclin:ab,ti,kw,tn OR supramycina:ab,ti,kw,tn OR tenutan:ab,ti,kw,tn OR tolexine:ab,ti,kw,tn OR torymycin:ab,ti,kw,tn OR tsurupioxin:ab,ti,kw,tn OR unidox:ab,ti,kw,tn OR veemycin:ab,ti,kw,tn OR viadoxin:ab,ti,kw,tn OR 'vibra s':ab,ti,kw,tn OR 'vibra tabs':ab,ti,kw,tn OR vibrabiotic:ab,ti,kw,tn OR vibracina:ab,ti,kw,tn OR vibradox:ab,ti,kw,tn OR vibramicina:ab,ti,kw,tn OR vibramycin:ab,ti,kw,tn OR vibraveineuse:ab,ti,kw,tn OR vibravenos:ab,ti,kw,tn OR vibravet:ab,ti,kw,tn OR 'viradoxyl‐n':ab,ti,kw,tn OR wanmycin:ab,ti,kw,tn OR zadorin:ab,ti,kw,tn OR zenavod:ab,ti,kw,tn OR '10592‐13‐9':ab,ti,kw,tn OR '17086‐28‐1':ab,ti,kw,tn OR '564‐25‐0':ab,ti,kw,tn OR '94088‐85‐4':ab,ti,kw,tn
#57 'minocycline'/exp
#58 minocyclin*:ab,ti,kw,tn OR '7 dimethylamino 6 demethyl 6 deoxytetracycline':ab,ti,kw,tn OR akamin:ab,ti,kw,tn OR aknemin:ab,ti,kw,tn OR 'akne puren':ab,ti,kw,tn OR 'aknin mino':ab,ti,kw,tn OR aknosan:ab,ti,kw,tn OR arestin:ab,ti,kw,tn OR blemix:ab,ti,kw,tn OR borymycin:ab,ti,kw,tn OR cyclimycin:ab,ti,kw,tn OR cyclomin:ab,ti,kw,tn OR cyclops:ab,ti,kw,tn OR cynomycin:ab,ti,kw,tn OR dentomycin:ab,ti,kw,tn OR dynacin:ab,ti,kw,tn OR 'icht oral':ab,ti,kw,tn OR klinomycin:ab,ti,kw,tn OR klinotab:ab,ti,kw,tn OR lederderm:ab,ti,kw,tn OR logryx:ab,ti,kw,tn OR menocycline:ab,ti,kw,tn OR mestacine:ab,ti,kw,tn OR micromycin:ab,ti,kw,tn OR minakne:ab,ti,kw,tn OR minaxen:ab,ti,kw,tn OR 'mino 50':ab,ti,kw,tn OR 'mino wolff':ab,ti,kw,tn OR 'minox 50':ab,ti,kw,tn OR minocin:ab,ti,kw,tn OR minoclin:ab,ti,kw,tn OR minoclir:ab,ti,kw,tn OR minocyn:ab,ti,kw,tn OR minogalen:ab,ti,kw,tn OR minoline:ab,ti,kw,tn OR minolira:ab,ti,kw,tn OR minolis:ab,ti,kw,tn OR minomax:ab,ti,kw,tn OR minomycin:ab,ti,kw,tn OR minoplus:ab,ti,kw,tn OR minotab:ab,ti,kw,tn OR mirosin:ab,ti,kw,tn OR mynocine:ab,ti,kw,tn OR romin:ab,ti,kw,tn OR skinocyclin:ab,ti,kw,tn OR solodyn:ab,ti,kw,tn OR spicline:ab,ti,kw,tn OR vectran:ab,ti,kw,tn OR vectrin:ab,ti,kw,tn OR ximino:ab,ti,kw,tn OR '10118‐90‐8':ab,ti,kw,tn OR '11006‐27‐2':ab,ti,kw,tn OR '13614‐98‐7':ab,ti,kw,tn
#59 'tetracycline'/exp
#60 Tetracyclin*:ab,ti,kw,tn
#61 #47 OR #48 OR #49 OR #50 OR #51 OR #52 OR #53 OR #54 OR #55 OR #56 OR #57 OR #58 OR #59 OR #60
#62 #46 AND #61
#63 #32 AND #62

Appendix 4. PubMed search strategy

#1 ((randomized controlled trial[pt]) OR (controlled clinical trial[pt]) OR (randomised[tiab] OR randomized[tiab]) OR (placebo[tiab]) OR (drug therapy[sh]) OR (randomly[tiab]) OR (trial[tiab]) OR (groups[tiab])) NOT (animals[mh] NOT humans[mh])
#2 blephariti*[tw]
#3 (demodicosis[tw] OR demodicidosis[tw] OR demodecosis[tw] OR "cylindrical dandruff"[tw])
#4 (demodex[tw] OR "d. folliculorum"[tw] OR "d. brevis"[tw])
#5 Blepharoconjunctivitis[tw]
#6 (ocular[tw] AND (rosacea[tw] OR mites[tw]))
#7 (Meibomian[tw] AND gland*[tw])
#8 (ocular[tw] AND gland*[tw])
#9 ((eye[tw] OR eyes[tw] OR eyelid*[tw] OR ocular[tw] OR lid[tw] OR lids[tw]) AND inflamm*[tw])
#10 ((eye[tw] OR eyes[tw] OR eyelid*[tw] OR ocular[tw] OR lid[tw] OR lids[tw]) AND infect*[tw])
#11 ((eye[tw] OR eyes[tw] OR eyelid*[tw] OR lid[tw] OR lids[tw]) AND seborrheic[tw])
#12 ((eye[tw] OR eyes[tw] OR eyelid*[tw] OR lid[tw] OR lids[tw]) AND staphylococcal[tw])
#13 #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12
#14 (antibiotic*[tw] OR anti biotic*[tw] OR antibacterial*[tw] OR anti bacterial*[tw])
#15 Erythromycin*[tw] OR abomacetin[tw] OR acneryne[tw] OR acnesol[tw] OR "akne mycin"[tw] OR "aknederm ery gel"[tw] OR aknemycin[tw] OR anamycin[tw] OR "bonac gel"[tw] OR "c solve 2"[tw] OR cliniderm[tw] OR deripil[tw] OR duraerythromycin[tw] OR "e mycin"[tw] OR "e base"[tw] OR "e glades"[tw] OR "e solve 2"[tw] OR emgel[tw] OR "emu v"[tw] OR "emu ve"[tw] OR emuvin[tw] OR emycin[tw] OR eriecu[tw] OR "erimycin t"[tw] OR eriprodin[tw] OR eritimix[tw] OR eritrex[tw] OR eritrocina[tw] OR eritromicina[tw] OR erixyl[tw] OR ermycin[tw] OR ermysin[tw] OR "ery maxin"[tw] OR "ery b"[tw] OR "ery diolan"[tw] OR "ery maxin"[tw] OR "ery tab"[tw] OR eryacne[tw] OR eryacnen[tw] OR eryc[tw] OR erycen[tw] OR erycette[tw] OR erycin[tw] OR erycinum[tw] OR eryderm[tw] OR erydermec[tw] OR erydermer[tw] OR eryfluid[tw] OR erygel[tw] OR eryhexal[tw] OR erymax[tw] OR erymaxin[tw] OR erymed[tw] OR erysafe[tw] OR erystrat[tw] OR erytab[tw] OR "eryth mycin"[tw] OR erythelan[tw] OR erythmycin[tw] OR erythomycin[tw] OR "erythra derm"[tw] OR erythran[tw] OR "erythro 200"[tw] OR "erythro teva"[tw] OR "erythro statin"[tw] OR erythrogan[tw] OR erythrogel[tw] OR erythrogran[tw] OR erythroguent[tw] OR erythromid[tw] OR erythromycine[tw] OR erythromycinum[tw] OR erythroteva[tw] OR erytop[tw] OR erytrarco[tw] OR erytrociclin[tw] OR etinycine[tw] OR etrolate[tw] OR etromycin[tw] OR ilocap[tw] OR ilocaps[tw] OR iloticina[tw] OR ilotycin[tw] OR "inderm gel"[tw] OR labocne[tw] OR latotryd[tw] OR lederpax[tw] OR mephamycin[tw] OR oftamolets[tw] OR pantodrin[tw] OR pantomycin[tw] OR pce[tw] OR pharyngocin[tw] OR primacine[tw] OR "r p mycin"[tw] OR robimycin[tw] OR romycin[tw] OR roymicin[tw] OR "rp mycin"[tw] OR rythocin[tw] OR "sans acne"[tw] OR sansac[tw] OR "skid gel e"[tw] OR staticin[tw] OR stiemycin[tw] OR stimycine[tw] OR "t stat"[tw] OR tstat[tw] OR theramycin[tw] OR "114‐07‐8"[tw] OR "70536‐18‐4"[tw]
#16 Azithromycin*[tw] OR aruzilina[tw] OR atizor[tw] OR azadose[tw] OR azasite[tw] OR azatril[tw] OR azenil[tw] OR azibiot[tw] OR azimin[tw] OR azithral[tw] OR azitrocin[tw] OR azitromax[tw] OR azitromicin[tw] OR azitromicina[tw] OR aziwok[tw] OR azomyne[tw] OR aztrin[tw] OR azydrop[tw] OR azyter[tw] OR azythromycin[tw] OR bazyt[tw] OR "cp 62933"[tw] OR "cp 62993"[tw] OR cp62933[tw] OR cp62993[tw] OR forcin[tw] OR goxal[tw] OR inedol[tw] OR infectoazit[tw] OR "isv 401"[tw] OR isv401[tw] OR kromicin[tw] OR macrozit[tw] OR mezatrin[tw] OR octavax[tw] OR ordipha[tw] OR ribotrex[tw] OR sumamed[tw] OR sunamed[tw] OR tobyl[tw] OR toraseptol[tw] OR tromix[tw] OR trozocina[tw] OR ultreon[tw] OR vinzam[tw] OR xithrone[tw] OR "xz 450"[tw] OR xz450[tw] OR zaret[tw] OR zarom[tw] OR zentavion[tw] OR zetamax[tw] OR zeto[tw] OR zibramax[tw] OR zifin[tw] OR zimericina[tw] OR zistic[tw] OR zithromax[tw] OR zithrox[tw] OR zitinn[tw] OR zitrim[tw] OR zitrobifan[tw] OR zitrocin[tw] OR zitromax[tw] OR zmax[tw] OR "117772‐70‐0"[tw] OR "121470‐24‐4"[tw] OR "83905‐01‐5"[tw]
#17 Macrolide*[tw] OR macrotetrolide*[tw]
#18 Doxycyclin*[tw] OR atridox[tw] OR "BMY 28689"[tw] OR BMY28689[tw] OR "BU 3839T"[tw] OR BU3839T[tw] OR "5 hydroxy 6 deoxytetracycline"[tw] OR "6 deoxy 5 hydroxytetracycline"[tw] OR "6 deoxy 5 oxytetracycline"[tw] OR "6 deoxyoxytetracycline"[tw] OR "6 desoxy 5 hydroxytetracycline"[tw] OR "6beta deoxyoxytetracycline"[tw] OR ado a[tw] OR "alpha 6 desoxy 5 oxytetracycline"[tw] OR "alpha 6 deoxyoxytetracycline"[tw] OR amfomycin[tw] OR atrax[tw] OR azudoxat[tw] OR bactimos[tw] OR brandolin[tw] OR base dillin[tw] OR bassano[tw] OR biocolon[tw] OR biodoxis[tw] OR ironmycin[tw] OR cloran[tw] OR cyclo/dox[tw] OR dentistry[tw] OR doxycycline[tw] OR "deoxymycin dispersal"[tw] OR deoxymykoin[tw] OR deoxyoxytetracycline[tw] OR "desoxy oxytetracycline"[tw] OR dedoxycycline[tw] OR doximycin[tw] OR oryx[tw] OR dosil[tw] OR dotur[tw] OR doxiciclina[tw] OR doxycycline[tw] OR doxa[tw] OR dox tet[tw] OR "doxi‐sergo"[tw] OR oxybiotic[tw] OR doxycycline[tw] OR doxiline[tw] OR moximed[tw] OR doximycin[tw] OR dioxin[tw] OR oxine[tw] OR doxycycline[tw] OR doxsie[tw] OR doxy[tw] OR oxybiotin[tw] OR doxycen[tw] OR doxychel[tw] OR doxycin[tw] OR "doxycydine monohydrate"[tw] OR doxylam[tw] OR toxylin[tw] OR doxymycine[tw] OR doxy puren[tw] OR doxytec[tw] OR oxytri[tw] OR dumoulin[tw] OR furacycline[tw] OR endoxan[tw] OR etifoxine[tw] OR new cyclin[tw] OR "gs 3065"[tw] OR hygromycin[tw] OR sobralene[tw] OR idocyklin[tw] OR idocyklin[tw] OR interdomain[tw] OR investing[tw] OR joygamycin[tw] OR ludox[tw] OR magrin[tw] OR medimycin[tw] OR meslin[tw] OR mildon[tw] OR miraculin[tw] OR monodon[tw] OR nordoy[tw] OR "nsc 56228"[tw] OR oracea[tw] OR paldimycin[tw] OR "pernox gel"[tw] OR periostat[tw] OR redox[tw] OR reumycin[tw] OR respidx[tw] OR proximycin[tw] OR serowoky[tw] OR pyridoxine[tw] OR servidoxyne[tw] OR sigadoxin[tw] OR silicon[tw] OR sigadoxin[tw] OR spano[tw] OR supracycline[tw] OR supramycin[tw] OR tenuta[tw] OR toalexin[tw] OR toromycin[tw] OR tsurumi toxin[tw] OR univox[tw] OR venemycin[tw] OR vipoxin[tw] OR "vibra s"[tw] OR "Vibra Tabs"[tw] OR vabra biotic[tw] OR vibrating[tw] OR vibrator[tw] OR vibramycin[tw] OR vibramycin[tw] OR vabra veineuse[tw] OR vibravenos[tw] OR vibravet[tw] OR "viradoxyl‐n"[tw] OR kanamycin[tw] OR zadori[tw] OR zenavod[tw] OR "10592‐13‐9"[tw] OR "17086‐28‐1"[tw] OR "564‐25‐0"[tw] OR "94088‐85‐4"[tw]
#19 Minocyclin*[tw] OR "7 dimethylamino 6 demethyl 6 deoxytetracycline"[tw] OR akamin[tw] OR aknemin[tw] OR "akne puren"[tw] OR "aknin mino"[tw] OR aknosan[tw] OR arestin[tw] OR blemix[tw] OR borymycin[tw] OR cyclimycin[tw] OR cyclomin[tw] OR cyclops[tw] OR cynomycin[tw] OR dentomycin[tw] OR dynacin[tw] OR "Icht Oral"[tw] OR klinomycin[tw] OR klinotab[tw] OR lederderm[tw] OR logryx[tw] OR menocycline[tw] OR mestacine[tw] OR micromycin[tw] OR minakne[tw] OR minaxen[tw] OR "mino 50"[tw] OR "mino wolff"[tw] OR "minox 50"[tw] OR minocin[tw] OR minoclin[tw] OR minoclir[tw] OR minocyn[tw] OR minogalen[tw] OR minoline[tw] OR minolira[tw] OR minolis[tw] OR minomax[tw] OR minomycin[tw] OR minoplus[tw] OR minotab[tw] OR mirosin[tw] OR mynocine[tw] OR romin[tw] OR skinocyclin[tw] OR solodyn[tw] OR spicline[tw] OR vectran[tw] OR vectrin[tw] OR ximino[tw] OR "10118‐90‐8"[tw] OR "11006‐27‐2"[tw] OR "13614‐98‐7"[tw]
#20 Tetracyclin*[tw]
#21 #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20
#22 #13 AND #21
#23 #1 AND #22
#24 Medline[sb]
#25 #23 NOT #24

Appendix 5. LILACS search strategy

(MH:C11.338.133$ OR blephariti$ OR blefariti$ OR blefarit$ OR demodicosis OR demodicidosis OR demodecosis OR "cylindrical dandruff" OR demodex OR "d. folliculorum" OR "d. brevis" OR blepharoconjunctivitis OR (ocular rosacea) OR (ocular mites) OR (meibomian gland$) OR (Glándulas Tarsale$) OR (Glândulas Tarsai$) OR MH:A09.371.337.614$ OR MH:A10.336.827.600$ OR (ocular gland$) OR (eye$ inflamm$) OR (ocular inflamm$) OR (eye$ infect$) OR (ocular infect$) OR (eye$ seborrheic) OR (eye$ staphylococcal)) AND (antibiotic$ OR "anti biotic" OR "anti biotics" OR "anti bacterial" OR "anti bacterials" OR Antibacteriano$ OR MH:D27.505.954.122.085$ OR MH:SP4.022.238.359$ OR MH:VS2.002.001.012$ OR MH:VS2.004.001.002.006$ OR Erythromycin$ OR Eritromicin$ OR MH:D02.540.576.500.992$ OR Tetracyclin$ OR Tetraciclina$ OR MH:D02.455.426.559.847.562.900$ OR MH:D04.615.562.900$ OR Doxycyclin$ OR Doxiciclina$ OR MH:D02.455.426.559.847.562.900.200$ OR MH:D04.615.562.900.200$ OR Azithromycin$ OR Azitromicina$ OR MH:D02.540.576.500.992.050$ OR Minocyclin$ OR Minociclina$ OR MH:D02.455.426.559.847.562.900.550$ OR MH:D04.615.562.900.550$ OR Macrolid$ OR MH:D02.540.505$ OR MH:D02.540.576.500$ OR MH:D04.345.674.500$)

Appendix 6. ClinicalTrials.gov search strategy

(blepharitis OR demodicosis OR demodicidosis OR demodecosis OR "cylindrical dandruff" OR demodex OR "d. folliculorum" OR "d. brevis" OR blepharoconjunctivitis OR meibomian gland OR ocular gland OR ocular rosacea OR ocular mites OR eye inflammation OR eye infection OR ocular inflammation OR ocular infection) AND (antibiotic OR antibacterial OR Erythromycin OR Tetracyclines OR Macrolides OR Doxycycline OR Azithromycin OR Minocycline)

Appendix 7. WHO ICTRP search strategy

blepharitis AND antibiotic OR blepharitis AND antibacterial OR blepharitis AND Erythromycin OR blepharitis AND Tetracycline OR blepharitis AND Doxycycline OR blepharitis AND Azithromycin OR blepharitis AND Minocycline OR blepharitis AND Macrolides
demodex AND antibiotic OR demodex AND antibacterial OR demodex AND Erythromycin OR demodex AND Tetracycline OR demodex AND Doxycycline OR demodex AND Azithromycin OR demodex AND Minocycline OR demodex AND Macrolides
blepharoconjunctivitis AND antibiotic OR blepharoconjunctivitis AND antibacterial OR blepharoconjunctivitis AND Erythromycin OR blepharoconjunctivitis AND Tetracycline OR blepharoconjunctivitis AND Doxycycline OR blepharoconjunctivitis AND Azithromycin OR blepharoconjunctivitis AND Minocycline OR blepharoconjunctivitis AND Macrolides
meibomian gland AND antibiotic OR meibomian gland AND antibacterial OR meibomian gland AND Erythromycin OR meibomian gland AND Tetracycline OR meibomian gland AND Doxycycline OR meibomian gland AND Azithromycin OR meibomian gland AND Minocycline OR meibomian gland AND Macrolides
eye infection AND antibiotic OR eye infection AND antibacterial OR eye infection AND Erythromycin OR eye infection AND Tetracycline OR eye infection AND Doxycycline OR eye infection AND Azithromycin OR eye infection AND Minocycline OR eye infection AND Macrolides
ocular inflammation AND antibiotic OR ocular inflammation AND antibacterial OR ocular inflammation AND Erythromycin OR ocular inflammation AND Tetracycline OR ocular inflammation AND Doxycycline OR ocular inflammation AND Azithromycin OR ocular inflammation AND Minocycline OR ocular inflammation AND Macrolides
ocular infection AND antibiotic OR ocular infection AND antibacterial OR ocular infection AND Erythromycin OR ocular infection AND Tetracycline OR ocular infection AND Doxycycline OR ocular infection AND Azithromycin OR ocular infection AND Minocycline OR ocular infection AND Macrolides

Data and analyses

Comparison 1. Oral antibiotics versus placebo.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Mean change in the Ocular Surface Disease Index at week 12	1		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
1.2 Mean change in bulbar conjunctival hyperemia at week 12	1		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
1.3 Mean number of symptoms at month 1	1		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
1.4 Subjective improvement in symptoms at month 1	1		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
1.5 Mean Schirmer's test score at month 1 (mm/5 min)	1		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
1.6 Tear film break‐up time at month 1 (seconds)	1		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
1.7 Adverse events	2		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only

Characteristics of studies

Characteristics of included studies [ordered by study ID]

NCT00560703.

Study characteristics
Methods	Study design: parallel‐group randomized controlled trial Number randomized (total and per group): 70 participants in total; 46 participants in oral antibiotics group and 24 participants in placebo group Unit of randomization (individual or eye): individual Number analyzed (total and per group): 70 participants in total; 46 participants in oral antibiotics group and 24 participants in placebo group Unit of analysis (individual or eye): individual Exclusions and losses to follow‐up (total and per group): 6 participants (2 excluded due to adverse events, 2 excluded due to protocol violation, 1 withdrawn by participant, 1 excluded due to non‐compliance) in placebo group How were missing data handled?: not reported Length of follow‐up: 12 weeks Reported power calculation (Y/N), if yes, sample size and power: Y, 80% "It was anticipated that the difference between the treatment groups in mean reduction from baseline in OSDI scores will be approximately 7 points. A pooled standard deviation of 9.0 for the mean change from baseline OSDI score is expected. Under those assumptions a total of approximately 63 evaluable patients (42 COL‐101 patients and 21 placebo patients) is sufficient to provide 80% power."
Participants	Country: USA Setting: 8 sites including tertiary care and private practice Baseline characteristics: 1. Oral antibiotics, n = 46 Age (mean ± SD, range): 54.5 ± 13.46 years Gender: 16 men and 30 women 2. Placebo, n = 24 Age (mean ± SD, range): 58.2 ± 14.27 years Gender: 11 men and 13 women Overall, n = 70 Age (mean ± SD, range): 55.7 ± 13.76 years Gender: 27 men and 43 women Inclusion criteria: 18 years of age and older all sexes blepharitis facial rosacea Exclusion criteria: pregnant or nursing women allergy to tetracyclines recent eye surgery past or current use of isotretinoin patients who are achlorhydric patients who have had gastric by‐pass surgery Baseline equivalence: not reported
Interventions	1. Oral antibiotics: oral 40 mg doxycyline once a day for 84 days 2. Placebo: sugar capsule once a day for 84 days
Outcomes	Primary outcome: bulbar conjunctival hyperemia, OSDI Secondary outcomes: Schirmer tear test, tear break‐up time, meibum character/fluidity, meibomian gland inspissation Adverse outcomes: Oral antibiotics: 1 (2.17%) participant migraine headache (severe adverse event), 5 (10.87%) participants headache Placebo: 1 (4.17%) participant non‐Hodgkin's lymphoma (severe adverse event) Measurement time points: 12 weeks Other issues with outcome assessment (e.g. quality control for outcomes, if any): none
Notes	Study period: November 2007 to July 2009 Publication language: data reported in trial registry only Trial registration:NCT00560703 Conflicts of interest: not reported Funding source: this study was funded by Galderma
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of random sequence generation was not reported.
Allocation concealment (selection bias)	Unclear risk	Allocation concealment before assignment was not reported.
Blinding of participants and personnel (performance bias)	Low risk	According to clinical trial registry, this study applied masking to participants, care provider, investigator and outcomes assessor; sugar capsules were used as placebo.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Clinical trial registry describes masking of outcome assessors to the assignment.
Incomplete outcome data (attrition bias) All outcomes	Low risk	There were no missing outcome data.
Selective reporting (reporting bias)	High risk	Secondary outcomes (Schirmer tear test, tear break‐up time, meibum character/fluidity, meibomian gland inspissation) presented in clinical registry were not reported on in the final results.

Yoo 2005.

Study characteristics
Methods	Study design: parallel‐group randomized controlled trial Number randomized (total and per group): 300 eyes of 150 participants in total; 100 eyes of 50 participants in each group Unit of randomization (individual or eye): individual Number analyzed (total and per group): 278 eyes of 139 participants in total; 92 eyes of 46 participants each in high‐ and low‐dose doxycycline groups, and 94 eyes of 47 participants in placebo group Unit of analysis (individual or eye): individual  Exclusions and losses to follow‐up (total and per group): 11 participants in total; 4 participants (4 excluded due to adverse events) in high‐dose group, 4 participants (3 excluded due to adverse events) in low‐dose group, 3 participants (1 excluded due to adverse events) in placebo group How were missing data handled?: excluded Length of follow‐up: 1 month Reported power calculation (Y/N), if yes, sample size and power: not reported
Participants	Country: South Korea Setting: Dankook University College of Medicine Baseline characteristics: 1. Oral antibiotics (high dose), n = 46 Age (mean ± SD, range): 47.93 ± 12.30 years Gender: 20 men and 26 women 2. Oral antibiotics (low dose), n = 46 Age (mean ± SD, range): 47.05 ± 12.24 years Gender: 16 men and 30 women 3. Placebo, n = 47 Age (mean ± SD, range): 46.67 ± 14.11 years Gender: 14 men and 33 women Overall, n = 150 (all randomized participants) Age (mean ± SD, range): 47.2 ± 12.36 years Gender: 55 men and 95 women Inclusion criteria: newly diagnosed with chronic meibomian gland dysfunction with grade 2 or worse meibomian gland destruction or meibomian gland orifice obstruction, and whose symptoms failed to improve despite warm compression, lid massage, lid scrub, and topical eye drops or ointment therapy for more than 2 months Exclusion criteria: not reported Baseline equivalence: baseline comparable
Interventions	1. Oral antibiotics (high dose): 200 mg of systemic doxycycline (doxycycline monohydrate, VIRADOXYL‐N, Jin‐Yang pharmacy) twice a days for 1 month 2. Oral antibiotics (low dose): 20 mg of systemic low dose doxycycline (doxycycline hyclate, DENTISTAR, Ha‐Na pharmacy) twice a day for 1 month 2. Placebo: placebo twice a day for 1 month
Outcomes	Primary outcome: Schirmer's test, tear break‐up time, subjective symptoms Secondary outcomes: not distinguished Adverse outcomes: Oral antibiotics (high dose): cardinal side effects (18 participants) withdrawn due to adverse events (4 participants) Oral antibiotics (low dose): cardinal side effects (8 participants) withdrawn due to adverse events (2 participants) Placebo: cardinal side effects (3 participants) withdrawn due to adverse events (1 participant) Others: gastrointestinal problems (i.e. dyspepsia, nausea, diarrhea, etc.) (21 participants) itchy skin, urticaria and erythematous papules (7 participants each) stomatitis (1 participant) Measurement time points: 1 month Other issues with outcome assessment (e.g. quality control for outcomes, if any): none
Notes	Study period: between January and December 2003 Publication language: English Trial registration: not identified Conflicts of interest: not reported Funding source: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of random sequence generation was not reported: "The patients were divided randomly into 3 groups according to dosage of doxycycline."
Allocation concealment (selection bias)	Unclear risk	Allocation concealment before assignment was not reported.
Blinding of participants and personnel (performance bias)	Low risk	"Both the nurses that distributed the medication and the patients were blinded to medication and treatment group"
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Masking of outcome assessors was not reported.
Incomplete outcome data (attrition bias) All outcomes	High risk	11 (7.3%) participants who were randomized were not included in the analysis.
Selective reporting (reporting bias)	Unclear risk	Neither protocol nor clinical registry was available.

OSDI: Ocular Surface Disease Index

SD: standard deviation

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Abboud 1972	Not an RCT
Al‐Hity 2016	Not an RCT
Aragones 1973	Not interventions of interest
Bakar 2009	Not an RCT
Bartholomew 1982	Not an RCT
De Benedetti 2019	Not interventions of interest
EUCTR2015‐002508‐94‐AT	Not interventions of interest
Filho 2011	Not an RCT
Frucht‐Pery 1989	Not an RCT
Gollnick 2010	Not participants of interest
Graham 1960	Not interventions of interest
Greene 2014	Not an RCT
Gumerova 2004	Not interventions of interest
Igami 2011	Not an RCT
JPRN‐UMIN000025159	Not interventions of interest
Lee 2012	Not an RCT
Margolis 1981	Not an RCT
Marmion 1969	Not an RCT
NCT00355459	Study was withdrawn before enrollment.
NCT00732446	Not interventions of interest
NCT01600625	Not an RCT
NCT01783860	Not interventions of interest
Park 1986	Not interventions of interest
Schaller 2019	Not participants of interest
Seal 1995	Cross‐over design
Souchier 2008	Not an RCT
TCTR20140524001	Not interventions of interest
TCTR20180810001	Not interventions of interest

RCT: randomized controlled trial

Characteristics of ongoing studies [ordered by study ID]

ChiCTR‐IPC‐15006715.

Study name	Azithromycin and doxycycline in the management of meibomian gland dysfunction
Methods	Parallel‐group randomized controlled trial
Participants	Inclusion criteria: symptomatic MGD not responding to warm compresses, lid massage, and lubricants older than 12 years able to give informed consent able to comply with treatment and follow‐up schedule Exclusion criteria: on treatment for MGD or dry eye syndrome other than warm compresses, lid massage, and lubricants in the past 3 months Schirmer's test without anesthesia < 5 mm per 5 minutes in both eyes—pretreatment LipiView score >= 100 in both eyes known allergy to tetracycline or macrolide pregnant or breastfeeding women women planning pregnancy hepatic impairment or on hepatotoxic drugs severe renal impairment ocular condition that causes similar symptoms and signs as MGD, including structural lid abnormalities, lagophthalmos, conjunctivochalasis, contact lens wear, thyroid eye disease on systemic medications that may contribute to dry eye without being on a stable dose in the past 1 month
Interventions	1. Oral antibiotics: oral doxycycline 100 mg (1 tablet) a day for 28 days; topical azithromycin to both eyes, twice daily for 3 days, then once daily for 25 days 2. Placebo: oral multivitamin (Daily 1 Plus) 1 tablet a day concurrently for 28 days as placebo; topical azithromycin to both eyes, twice daily for 3 days, then once daily for 25 days
Outcomes	Primary outcome: improvement in symptom frequency score based on self‐administered OSDI questionnaire at month 1 Secondary outcomes: improvement in symptom frequency score based on self‐administered OSDI questionnaire at month 3 improvement in symptom severity score based on self‐administered SPEED questionnaire clinical grading of MGD based on conjunctival injection, lid margin redness, ocular surface staining, meibomian gland expressability, number of glands plugged and quality of meibum13 best‐corrected visual acuity lipid layer thickness measurement by LipiView interferometer (TearScience Inc; Morrisville, NC, USA) LipiView interferometer (TearScience Inc; Morrisville, NC, USA) measures interferometric color units (ICU), where 1 ICU approximately reflects 1 nm of the lipid layer thickness.18 A LipiView image of the tear film can be captured during a non‐invasive in‐office exam that takes about 5 minutes. The tear film can be seen as an array of colors called specular observations that are reflected when the participant rests his head on the chin‐rest and a controlled light source is directed towards the front surface of the eye. non‐invasive tear break‐up time, infrared meibography, tear film assessment and R‐scan (bulbar conjunctival redness evaluation) with OCULUS Keratograph 5M (Oculus, Wetzlar, Germany) OCULUS Keratograph 5M (Oculus, Wetzlar, Germany) uses white or infrared illumination to assess the tear film and tear break‐up time; infrared camera for meibomian gland imaging; and color camera for conjunctival blood vessel imaging. All measurements will be non‐invasive. Participants will be instructed to rest head on the chin‐ rest during imaging, which takes about 5 minutes. tear analysis (10 participants in each arm): osmolarity, cytokines, matrix metalloproteinases side effects noted during treatment compliance to treatment Length of follow‐up: 3 months
Starting date	18 July 2015 (anticipated trial start date)
Contact information	Dr Vanissa Wing‐see Chow, vanissa.chow@gmail.com
Notes

NCT03953118.

Study name	Azithromycin for meibomian gland disease
Methods	Parallel‐group randomized controlled trial
Participants	Inclusion criteria: symptomatic meibomian gland disease, defined as patient‐reported ocular surface symptoms such as dryness, grittiness, foreign body sensation, or eye fatigue in combination with clinically identifiable meibomian gland disease with grade 2 or greater involvement on the Meibomian Gland Grading Scale (Section XV, item 2) OSDI score ≥ 20 ability to give informed consent Exclusion criteria: age less than 18 years allergy or intolerance to oral azithromycin or topical dexamethasone allergy or intolerance to the preservatives used in topical ophthalmic 0.1% dexamethasone: sodium bisulfite, phenylethyl alcohol, benzalkonium chloride history of prolonged QT interval, history of torsades des pointes, congenital long QT syndrome, bradyarrhythmias, heart failure patients currently taking medications that prolong the QT interval aqueous deficiency dry eye defined as Schirmer's strip testing without anesthesia with < 5 mm of tears on 2 separate tests ocular surface inflammatory disease, including cicatrizing conjunctivitis, graft versus host disease, Stevens‐Johnson syndrome atopic disease with ocular involvement limbal stem cell deficiency oral or topical ophthalmic antibiotic use within the last 90 days oral prednisone use > 5 mg per day topical ophthalmic steroid use within the past 30 days topical ophthalmic anti‐inflammatory (including non‐steroidal anti‐inflammatory medications, lifitegrast, or ciclosporin) use within the past 30 days patients who are currently pregnant, planning on becoming pregnant during the study period, or currently breastfeeding
Interventions	1. Oral azithromycin (Zithromax) dosed at 1 g per week for 3 weeks 2. Oral placebo tablet
Outcomes	Primary outcome: OSDI questionnaire Secondary outcomes: ocular surface microbiome testing Dry Eye Questionnaire‐5 Neuropathic Pain Inventory for the Eye Personal Health Questionnaire‐9 Length of follow‐up: 3 months
Starting date	1 November 2019 (estimated study start date), 1 December 2019 (estimated study completion date)
Contact information	Dr Julie Schallhorn, julie.schallhorn@ucsf.edu
Notes

MGD: meibomian gland dysfunction

OSDI: Ocular Surface Disease Index

Differences between protocol and review

We excluded other biases (other potential threats to validity) in accordance with the updated Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019).

Contributions of authors

Conception and design of study: all review authors

Assessing the eligibility of relevant studies: NO, SN

Critically appraising risk of bias and extracting data: NO, SN

Conducting qualitative data synthesis: NO, SN

Data entry: NO, SN

Drafting the review or commenting on it critically for intellectual content: NO, SN, MSH

Final approval of the document to be published: all review authors

Sources of support

Internal sources

• No sources of support provided

External sources

• Cochrane Eyes and Vision US Project, supported by grant 1 U01 EY020522, National Eye Institute, National Institutes of Health, USA

• National Institute for Health Research (NIHR), UK

  • Richard Wormald, Co‐ordinating Editor for the Cochrane Eyes and Vision (CEV), received financial support for his CEV research sessions from the Department of Health through the award made by the National Institute for Health Research (NIHR) to Moorfields Eye Hospital NHS Foundation Trust and University College London Institute of Ophthalmology for a Specialist Biomedical Research Centre for Ophthalmology.

  • The NIHR also funded the CEV Editorial Base in London during the time this review was being written.

  The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the Systematic Reviews Programme, NIHR, NHS, or the Department of Health.

• Queen's University Belfast, UK

  Gianni Virgili, Co‐ordinating Editor for Cochrane Eyes and Vision’s work is funded by the Centre for Public Health, Queen’s University of Belfast, Northern Ireland 

Declarations of interest

Neil Onghanseng: none known
Sueko M Ng: none known
Muhammad Sohail Halim: none known
Quan Dong Nguyen: none known

New