Can the choice of artificial tears harm patients? A narrative review with an overview of the Nordic market

Abstract

Purpose

To provide an overview of artificial tears marketed in the following Nordic countries: Denmark, Finland, Norway, Iceland and Sweden. Furthermore, this review aimed to highlight the different preservatives and other constituents found in artificial tears in the Nordic market, focussing on adverse effects.

Methods

Artificial tears appearing in online pharmacies in Denmark, Norway, Sweden, Finland and Iceland were included, and tables listing their components were created. Based on the preservatives and other constituents found in the artificial tears on the Nordic market, a literature search was conducted to investigate differences and potential adverse effects. This was achieved using PubMed with MeSH and free text search terms. Included articles were (i) studies performed on humans in vivo or in vitro human models, (ii) published between 2000 and 2023 and (iii) focussing on adverse effects, preservatives, toxicity, other constituents and preservative‐free artificial tears.

Results

A total of 88 artificial tears were found on the Nordic market. Approximately 32% (28 out of 88) of the artificial tears contain preservatives. Eleven of these are preserved with benzalkonium chloride (BAK) or cetrimide. After a thorough literature search, evidence has been found for the toxic effects of BAK and cetrimide. There is no evidence of toxic effects of lubricants, osmoprotectants and lipids.

Conclusions

The findings of this review highlight the paradoxical fact that many treatments for ocular surface diseases contain toxic preservatives that may lead to worsening of the condition. In particular, long‐time toxic effects of BAK may contribute to disease progression.

1. INTRODUCTION

Artificial tears are used to treat a variety of conditions including ocular surface diseases that encompass dry eye disease (DED). According to the Tear Film & Ocular Surface Society (TFOS) Dry Eye Workshop II (DEWS II), dry eye disease is defined as ‘a multifactorial disease of the ocular surface characterized by a loss of homeostasis of the tear film, and accompanied by ocular symptoms, in which tear film instability and hyperosmolarity, ocular surface inflammation and damage, and neurosensory abnormalities play etiological roles’ (Craig et al., 2017) The overall prevalence of DED varies between 5 and 50% worldwide (Hedengran et al., 2022; Stapleton et al., 2017; Walsh & Jones, 2019). The incidence is highest in women, and increasing age is the leading risk factor for the development of DED (Heegaard et al., 2022). According to global sales, artificial tears accounted for $2.2 billion in 2018, and their market share continues to increase (Kathuria et al., 2021).

A primary goal of DED treatment is to restore tear film homeostasis. Artificial tears provide the necessary lubrication of the eye, reduce tear evaporation and stabilize the tear film (Pucker et al., 2016). It is associated with patient‐reported symptom improvement (Pucker et al., 2016; Walsh & Jones, 2019). Artificial tears vary in terms of viscosity, composition, pH, electrolytes, osmolarity and the type of preservatives utilized. Preservatives can affect the ability to protect and restore the ocular surface and thus the tolerability for DED patients (Kathuria et al., 2021; Tong et al., 2012). Artificial tears lack most biologically active components found in natural tears, such as specific anti‐inflammatory proteins, lactoferrin, lysozymes, immunoglobulin A and lipid‐binding proteins (Pucker et al., 2016).

Most artificial tears consist of lubricants. These agents improve tear film thickness, protect against external stressors and increase water retention time preventing water loss (Kathuria et al., 2021; Labetoulle, Benitez‐Del‐Castillo, et al., 2022). Lubricants are known to be mucoadhesive and mucomimetic (Labetoulle, Benitez‐Del‐Castillo, et al., 2022). Therefore, they at least partially help maintain the mucin layer in the case of DED due to mucin deficiency (Labetoulle, Benitez‐Del‐Castillo, et al., 2022). Lipids (emollients) are also added to some formulations. Lipids replenish the lipid layer of the tear film, which is important to maintain homeostasis of the ocular surface and reduce evaporation of the underlying aqueous layer (Labetoulle, Benitez‐Del‐Castillo, et al., 2022).

Tear film hyperosmolarity plays a significant role in DED. Hyperosmolarity of the tear film causes oxidative stress and stimulates a cascade of inflammatory events involving NF‐kB signalling pathways and the generation of inflammatory mediators (Bron et al., 2017). Hyperosmolarity is the main factor causing damage to the epithelial and goblet cells, leading to further tear film instability (Heegaard et al., 2022). Naturally occurring electrolytes are important for maintaining the osmotic and thus the homeostatic balance of the ocular surface. Electrolytes added to artificial tears are used to mimic the electrolyte profile of a healthy tear film. Therefore, hypotonic artificial tears can minimize the symptoms of DED (Labetoulle, Benitez‐Del‐Castillo, et al., 2022).

Free radical scavengers such as vitamin A and vitamin E are used in some artificial tears to reduce oxidative stress (Labetoulle, Benitez‐Del‐Castillo, et al., 2022). Artificial tears contain a variety of buffers such as citrate, phosphate and borate buffers to control the pH value.

Another important component of artificial tears are osmoprotectants, which are used to prevent oxidative stress, metalloproteinase synthesis and apoptosis of corneal and conjunctival epithelial cells due to hyperosmolarity of the tear film (Kathuria et al., 2021; Labetoulle, Benitez‐Del‐Castillo, et al., 2022). In vitro studies with cultured human corneal epithelial cells show that osmoprotectants are absorbed by dehydrated cells, restoring their volume and preventing protein denaturation (Kathuria et al., 2021).

To prevent microbial growth, maintain sterility and thereby extend shelf life, preservatives are frequently used in multi‐dose units of artificial tears.

In summary, restoring tear film homeostasis is the goal of DED treatment, and the first‐line intervention is tear replacement (Heegaard et al., 2022; Kathuria et al., 2021; Pucker et al., 2016; Tong et al., 2012; Walsh & Jones, 2019). Many artificial tears are available over the counter and do not require a prescription, yet the choice of artificial tears can be critical for the patient.

In this review, we aim to investigate artificial tears in the Nordic countries: Denmark, Finland, Norway, Sweden and Iceland (when henceforth mentioning the ‘Nordic market’, we are referring to these five countries). The artificial tears will be highlighted in tables according to their constituents and preservatives. Furthermore, the preservatives and constituents found will be assessed in terms of adverse effects.

2. METHODS

Different artificial tears sold in the Nordic Market have been found using online pharmacies in each country. The different online pharmacies used were: apotek1.no, apoteket.se, apotea.se, mediapteekki.fi, webapoteket.dk, apopro.dk, netverslun.lyfja.is. Besides this, an overview of marketed artificial tears in the Nordic countries were obtained from Théa Pharma to identify further artificial tears. Information on the content of the specific artificial tears was obtained from the package leaflet. In cases, where the package leaflet was not available, the information was found on the manufacturer's website. The different artificial tears were categorized according to whether they were preserved or not. Furthermore, preserved artificial tears were also categorized according to their different preservatives. The ingredients were grouped into lubricants, osmoprotectants, lipids, electrolytes, preservatives and other constituents. Water was not categorized as an ingredient, as it is a component of all the artificial tears besides EvoTears®. Glycerin and glycerol were considered the same.

Based on the preservatives and other constituents found in the Nordic market, a literature search of their toxicity and adverse effects was conducted. The literature search was performed in the Medline database (PubMed) using controlled vocabulary with Medical Subject Headings (MeSH) and free text search terms. The search was limited to publications written in English, and studies conducted on humans in vivo or in vitro. Furthermore, a date restriction was applied, and articles before the year 2000 were not used. Studies focussing on constituents in ophthalmic solutions, toxicity and preservative‐free artificial tears are included. As there are limited studies on artificial tears, a few articles regarding glaucoma medications and their preservatives were included.

3. RESULTS

3.1. Constituents in artificial tears on the Nordic market

Artificial tears comprise up to six different component categories: lubricants, osmoprotectants, electrolytes, lipids, preservatives and other constituents (Table 1). Table 1 summarizes the different constituents and their frequency in the artificial tears in the Nordics. A total of 88 artificial tears were identified (Tables 2 and 3). Sixty (~68%) were preservative‐free, while 28 (~32%) contained at least one preservative, including benzalkonium chloride (BAK), cetrimide (CET), polyquaternium‐1 (PQ‐1), polyhexamethylene biguanide (PHMB), phenoxyethanol and stabilized oxychloro complex (SOC) encompassing Purite®, OcuPure® or Oxyd®. Five, six, seven and six artificial tears contained BAK, CET, SOCs and PQ‐1, respectively (Tables 2 and 3). PHMB was found in three artificial tear products, and the preservative phenoxyethanol was only found in one artificial tear. All artificial tears but seven contained lubricants with either carbomer, carboxymethylcellulose (CMC), dextran, glycerin/glycerol, hypromellose, hyaluronate, hyaluronic acid (HA), hydroxypropyl‐guar (HP‐guar), polyethylene glycol (PEG), polyvinyl alcohol, polysorbate, povidone, propylene glycol, sodium hyaluronate (SH), sorbitan monooleate, sorbitan oleate and TS‐polysaccharide. The most frequently used were SH and HA. A total of 23 (~26%) artificial tears contained various lipids. These include castor oil, cetalkonium chloride, lanolin oil, liposomes, LipozonEye, mineral oil, oleic acid, paraffin oil, perilla seed extract, petroleum jelly/vaseline, phosphatidylglycerol, phospholipid, sacha inchi seed oil, sea buckthorn oil, sea buckthorn seed oil sorbitan tristearate, soybean oil, soy lecithin and triglycerides (Tables 1, 2, 3).

TABLE 1.

An overview of the constituents of artificial tears and the number of different artificial tear products of the Nordic market containing lubricants, osmoprotectants, lipids, electrolytes, preservatives or others. The names of the different constituents are expressed as provided in the product information sources.

	Components	Number of artificial tear products (n = 88)
Lubricants	Carbomer, CMC, dextran 70, glycerin/glycerol, hypromellose, Hyaluronate, hyaluronic acid/hyaluronan, HP‐guar, PEG 300/400/8000, polyvinyl alcohol, polysorbate 80 povidone, propylene glycol, SH, sorbitan monooleate, sorbitan oleate, and TS‐polysaccharide	81 (~92%)
Osmoprotectants	Ectoine, erythritol, levocarnitine, sorbitol and trehalose	33 (~38%)
Lipids	Castor oil, cetalkonium chloride, lanolin oil, liposomes, LipozonEye, mineral oil, oleic acid, paraffin oil, perilla seed extract, petroleum jelly/vaseline, phosphatidylglycerol, phospholipid, sacha inchi seed oil, sea buckthorn oil, sea buckthorn seed oil, sorbitan tristearate, soybean oil, soy lecithin, and triglycerides	23 (~26%)
Electrolytes and Buffers	Aminomethylpropanol, boric acid, calcium chloride, calcium chloride dihydrate, citrate buffer, citric acid anhydrous, citric acid monohydrate, disodium edetate, disodium edetate dihydrate, disodium hydrogen phosphate, disodium phosphate, disodium phosphate dihydrate, disodium phosphate dodecahydrate, disodium tetraborate decahydrate, hydrochloric acid, magnesium chloride, magnesium chloride hexahydrate, monosodium phospohate, phosphate buffer, potassium chloride, potassium hydrogen phosphate, sodium acetate trihydrate, sodium bicarbonate, sodium borate, sodium borate decahydrate, sodium chloride, sodium citrate, sodium citrate dihydrate, sodium dihydrogen phosphate dihydrate, sodium edetate, sodium hydrogen phosphate, sodium hydroxide, sodium lactate, sodium phosphate, sodium tetraborate decahydrate, tris hydrochloride trometamol and zinc chloride	80 (~91%)
Preservatives	BAK, CET, OcuPure®, Oxyd®, phenoxyethanol, PHMB, PQ‐1 and Purite®	28 (~32%)
Others	Actinoquinol, aloe vera gel, blueberry extract liquid dexpanthenol, ethanol, euphradia officinalis, lysine monohydrate, mannitol, perfluorohexyloctane, poloxamer 188, polyocyl 40 stearate, tyloxapol, vitamin A, palmitate (retinol palmitate) and vitamin E/vitamin E tocopherol	27 (~31%)

Abbreviations: BAK, benzalkonium chloride; CET, cetrimide; CMC, carboxymethylcellulose; HA, hyaluronic acid; HP‐guar, hydroxypropyl‐guar; PEG, polyethylene glycol; PHMB, Polyhexanide methylbiguanide; PQ‐1, polyquaternium‐1; SH, sodium hyaluronat; TS‐polysaccharide, tamarind seed polysaccharide.

TABLE 2.

Overview of the preserved artificial tears product brands in the Nordic market listing brand name, lubricants, osmoprotectants, lipids, electrolytes, preservatives and others. The names of the different constituents are expressed as provided in the product information sources.

Preserved artificial tears
Brand name	Lubricants	Osmoprotectants	Lipids	Electrolytes and buffers	Preservatives	Others
Artificial tears preserved with benzalkonium chloride
Clear eyes Bright and Moist Eye Drops	Hypromellose			Boric acid, sodium borate, disodium edetate	BAK
Lacril®	Polyvinyl alcohol			Disodium edetate, sodium phosphate, sodium chloride	BAK
Oculac® (bottle)	Povidone			Sodium chloride, sodium lactate, potassium chloride, calcium chloride, magnesium chloride, boric acid, sodium hydroxide	BAK
Oftagel® (bottle)	Carbomer Polyvinyl alcohol	Sorbitol		Sodium acetate trihydrate	BAK	Lysine monohydrate
Viscous eye drops ‘Optha’®	Hypromellose			Sodium chloride	BAK
Artificial tears preserved with cetrimide
Artelac® (bottle)	Hypromellose	Sorbitol		Disodium phosphate dodecahydrate, sodium dihydrogen phosphate dihydrate, disodium edetate	CET
Bevita eye gel	Carbomer	Sorbitol		Sodium hydroxide, disodium edetate	CET
Lakrimont®	Carbomer	Sorbitol		Sodium hydroxide, disodium edetate	CET
Oxyal® Care Gel	Carbomer			Sodium hydroxide, disodium edetate	CET	Dexpanthenol (vitamin B5)
Visc‐ophtal® øjengel	Carbomer	Sorbitol		Sodium hydroxide	CET
Viscotears® (tube)	Carbomer	Sorbitol		Sodium hydroxide	CET
Artificial tears preserved with PQ‐1
Systane® Balance	HP‐guar Propylene glycol	Sorbitol	Mineral oil Dimyristoyl phosphatidylglycerol Sorbitan tristearate	Boric acid, disodium edetate	PQ‐1	Polyocyl 40 stearate
Systane® Complete	HP‐guar Propylene glycol	Sorbitol	Mineral oil Dimyristoyl phosphatidylglycerol Sorbitan tristearate	Boric acid, hydrochloric acid and/or sodium hydroxide, disodium edetate	PQ‐1	Polyocyl 40 stearate
Systane® Gel drops	HP‐guar PEG 400 Propylene glycol	Sorbitol		Sodium chloride, potassium chloride, boric acid, aminomethylpropanol, disodium edetate	PQ‐1
Systane® Hydration	HP‐guar PEG 400 Propylene glycol Sodium Hyaluronate	Sorbitol		Sodium chloride, sodium citrate, potassium chloride, boric acid, sodium borate, aminomethylpropanol, disodium edetate	PQ‐1
Systane® Ultra	HP‐guar PEG 400 Propylene glycol	Sorbitol		Sodium chloride, potassium chloride, boric acid, aminomethylpropanol	PQ‐1
Tears Naturale®	Dextran 70 Hypromellose			Sodium tetraborate decahydrate, sodium chloride, potassium chloride, sodium hydroxide	PQ‐1
Artificial tears preserved with stabilized oxychloro complexes
Blink® Contacts	Sodium hyaluronate			Sodium chloride, potassium chloride, calcium chloride, magnesium chloride, boric acid	OcuPure®
Blink® Intensive Tears (bottle)	Sodium hyaluronate PEG 400			Sodium chloride, potassium chloride, calcium chloride, magnesium chloride, boric acid, sodium borate	OcuPure®
Blink® intensive Tears Plus	Sodium hyaluronate PEG 400			Sodium chloride, potassium chloride, calcium chloride, magnesium chloride, boric acid, sodium borate	OcuPure®
Optive®	CMC 0.5% Glycerin 0.9%	Levocarnitine Erythritol		Potassium chloride, calcium chloride dihydrate, magnesium chloride hexahydrate, boric acid, sodium borate decahydrate, sodium citrate dihydrate	Purite®
Optive® Fusion	CMC 0.5% Hyaluronate (HA) Glycerin 0.9%	Erythritol		Potassium chloride, calcium chloride dihydrate, sodium citrate dihydrate, magnesium chloride hexahydrate, Boric acid, sodium borate decahydrate	Purite®
Optive® Plus	CMC 0.5% Glycerin 1.0% Polysorbate 80	Levocarnitine Erythritol	Castor oil		Purite®
Oxyal®	Sodium hyaluronate PEG 8000			Sodium chloride, potassium chloride, calcium chloride dihydrate, magnesium chloride hexahydrate, boric acid	Oxyd®
Artificial tears preserved with other preservatives
Blink® Intensive Triple Action	Sodium hyaluronate PEG 400		Liposomes	Disodium edetate	PHMB	Vitamin E tocopherol
Desodrop®	Hypromellose		LipozonEye (ozonized sunflower oil, soy phospholipids)	Boric acid, sodium tetraborate decahydrate, sodium edetate	PHMB
Ocutein Sensitive plus	Sodium hyaluronate 0.10%			Disodium edetate	PHMB	Blueberry extract liquid
Tearsagain® lipidspray			Soy lectithin	Sodium chloride	Phenoxyethanol	Ethanol, vitamin A palmitate, vitamin E

Note: Glycerin and glycerol are considered the same.

Abbreviations: BAK, benzalkonium chloride; CET, cetrimide; CMC, carboxymethylcellulose; HP‐guar, hydroxypropyl‐guar; PEG, polyethylene glycol; PHMB, Polyhexanide methylbiguanide; PQ‐1, polyquaternium‐1.

TABLE 3.

Overview of the preservative‐free artificial tears product brands in the Nordic market listing brand name, lubricants, osmoprotectants, lipids, electrolytes and others. The names of the different constituents are expressed as provided in the product information sources.

Preservative‐free artificial tears
Brand name	Lubricants	Osmoprotectants	Lipids	Electrolytes and buffers	Others
A. Vogel	Hyaluronic acid			Sodium chloride, boric acid, disodium tetraborate decahydrate, sodium hydroxide, hydrochloric acid	Euphrasia officinalis
Apoteket Ögondroppar	Sodium hyaluronate	Sorbitol		Sodium chloride, potassium chloride, calcium chloride, magnesium chloride, citrate buffer
Artelac®	Hypromellose	Sorbitol		Disodium phosphate dodecahydrate, sodium dihydrogen phosphate dihydrate
Bepanthen Eye drops	Sodium hyaluronate			Sodium chloride, sodium citrate, citric acid	Dexpanthenol
Bepanthen intensive Eye drops	Sodium hyaluronate Carbomer		Phospholipids, Oleic acid soybean oil		Dexpanthenol, vitamin A, vitamin E
Bevita® eye	Hyaluronic acid			Sodium citrate, citric acid monohydrate, sodium chloride
Bevita® eye voiteleva silmavoide			Liquid paraffin Petroleum jelly		Retinol palmitate (vitamin A)
Bevita® eye pro	Hyaluronic acid TS‐polysaccharide			Sodium citrate, citric acid monohydrat	Mannitol
Bevita® Eye Silmäsuike	Glycerin	Trehalose	Sacha inchi seed oil	Tromethamine/citrate emulsion	Dexpanthenol
Blink® Intensive Tears	Sodium hyaluronate PEG 400			Sodium chloride, potassium chloride, calcium chloride dihydrate, magnesium chloride, boric acid, sodium borate
Cationorm®	Glycerin		Paraffin oil Cetalkonium chloride	Tromethamine, Tris hydrochloride	Surfactant: Tyloxapol Poloxamer 188
Celluvisc®	CMC 1.0%			Sodium chloride, sodium lactate, potassium chloride, calcium chloride
Cellufluid®	CMC 0.5%			Sodium chloride, sodium lactate, potassium chloride, calcium chloride dihydrate, magnesium chloride hexahydrate, sodium hydroxide or hydrochloric acid
Clarieye®	Sodium hyaluronate			Sodium chloride, sodium citrate, citric acid	Dexpanthenol
EvoTears®					Perfluorohexyloctane
Hyabak®	Sodium hyaluronate			Sodium chloride, hydrochloric acid, trometamol	Actinoquinol
Hylo‐Care®	Sodium hyaluronate			Citric acid anhydrous, sodium Citrate	Dexpanthenol
Hylo‐Comod®	Sodium hyaluronate	Sorbitol		Citrate buffer
Hylo‐Dual Care®	Sodium hyaluronate			Sodium citrate	Dexpanthenol
Hylo‐Dual Intense®	Sodium hyaluronate	Ectoine		Boric acid, sodium tetraborate decahydrate
Hylo‐Fresh®	Sodium Hyaluronate			Boric acid, sodium tetraborate decahydrate	Euphrasia Officinalis
Hylo‐Gel®	Sodium hyaluronate	Sorbitol		Citrate buffer
Hylo‐Night®			Liquid paraffin oil Lanolin oil Petroleum jelly		Retinol palmitate (vitamin A)
Hyprosan®	Hypromellose Sodium hyaluronate			Sodium chloride Disodium phosphate dodecahydrate, sodium dihydrogen phosphate dihydrate
Iiris eye drops	Sodium hyaluronate			Hydrochloric acid, sodium chloride, trometamol
Membrasin® vision eye spray	Sodium Hyaluronate Glycerol Polysorbate 80 Sorbitan monooleate		Sea buckthorn oil	Citric acid monohydrate, potassium chloride, sodium chloride, Trometamol
Metode øyedråber	Hyaluronic acid 0.2%			Disodium edetate
Metode ekstra tørre øyne	Hyaluronic acid 0.3%			Disodium edetate
Metode ekstra tørre øyne with aloe vera	Hyaluronic acid 0.3%			Disodium edetate	Aloe vera gel
Oculac®	Povidone			Sodium chloride, sodium lactate, potassium chloride, calcium chloride, magnesium chloride, boric acid, sodium hydroxide
Ocutears® Hydro+	Sodium hyaluronate 0.2%			Boric acid, disodium tetraborate decahydrate, sodium chloride
Oftagel®	Carbomer Polyvinyl alcohol	Sorbitol		Sodium acetate trihydrate	Lysine monohydrate
Ophtim Eye® Forte	Sodium hyaluronate			Sodium chloride, monosodium phosphate, disodium phosphate
Ophtim Eye® Hydra	Sodium Hyaluronate 0.2%			Disodium phosphate, monosodium phosphate, sodium chloride
Oxyal® Total Care Spray	Glycerin Polysorbate 80 Sorbitan oleate	Trehalose	Sacha inchi seed oil	Trometamol/citrate buffer emulsion	Dexpanthenol
Oxyal® Triple Action	Sodium hyaluronate Carbomer Glycerin		Triglycerides	Sodium hydroxide
Oxyal® Triple Action Spray	Sodium hyaluronate, Carbomer Glycerin		Triglycerides	Sodium hydroxide
Oxyal® Trehalos Duo Action	Sodium hyaluronate	Trehalose		Phosphate buffer
Oxyal® Trehalos Tripple Action	Sodium hyaluronate Glycerin	Trehalose	Sacha inchi seed oil	Hypotonic trometamol/citrate buffer emulsion
Piiloset Biodrop MD® plus	Sodium hyaluronate	Trehalose		Low phosphate buffer solution
Puro™ Suoja	High‐molecular‐weight hyaluronan	Trehalose	Sacha inchi seed oil	Trometamol/citrate buffer emulsion
Puro™ Teho	Sodium hyaluronate	Trehalose		Low phosphate buffer solution
Puro™ Tyrni	Hyaluronic acid Glycerin Polysorbate 80 Sorbitan Monooleate		Sea buckthorn seed oil	Trometamol/citrate buffer emulsion, potassium chloride
Puro™ Vire	Hyaluronan 0.2%			Sodium chloride, disodium hydrogen phosphate, sodium dihydrogen phosphate
Systane® Hydration	HP‐guar PEG 400 Propylene glycol Sodium hyaluronate	Sorbitol		Sodium chloride, potassium chloride, sodium borate, boric acid, hydrochlorid acid/and/or sodium hydroxide, aminomethylpropanol
Systane® Ultra	HP‐guar PEG 400 Propylene glycol	Sorbitol		Sodium chloride, potassium chloride, boric acid, hydrochlorid acid and/or sodium hydroxide, aminomethylpropanol
Tearsagain® Eye drops	Sodium hyaluronate			Sodium chloride, potassium chloride, calcium chloride, magnesium chloride, sodium bicarbonate, Sodium hydrogen phosphate
Tearsagain® Sensitive			Soy lectithin	Sodium chloride	Ethanol, Dexpanthenol Vitamin A palmitat, Vitamin E
Thealipid®	Glycerin		Phospholipid Triglyceride Soybean oil	Disodium edetate	Vitamin E (a‐Tocopherol)
Thealoz® Duo	Sodium hyaluronate	Trehalose		Hydrochloric acid, sodium chloride, Trometamol
Thealoz® Duo Gel	Sodium hyaluronate Carbomer	Trehalose Sorbitol		Sodium hydroxide
Torra ögon apofri	Sodium hyaluronate	Sorbitol		Sodium hydroxide, hydrochloric acid
Viscotears®	Carbomer	Sorbitol		Sodium hydroxide
Vismed artificial tears	Sodium hyaluronate			Sodium chloride, potassium chloride, disodium phosphate, sodium citrate, magnesium chloride, calcium chloride
VisMed artificial tears gel	Sodium hyaluronate			Sodium chloride, potassium chloride, disodium phosphate, sodium citrate, magnesium chloride, calcium chloride
VisMed multi	Sodium hyaluronate			Sodium chloride, potassium chloride, disodium phosphate, sodium citrate, magnesium chloride, calcium chloride
Vita‐Pos® Eye Ointment			Liquid paraffin oil, lanolin oil, vaseline		Retinyl palmitate (vitamin A)
Zalispray®	Sodium hyaluronate PEG 300		Liposomes Perilla seed extract	Sodium chloride, sodium tetraborate decahydrate, boric acid, disodium edetate dihydrate
Zilkeye™	Sodium hyaluronate			Sodium chloride, potassium hydrogen phosphate, disodium phosphate dihydrate
Øjensalve Neutral ‘Ophtha’®			Liquid paraffin oil, vaseline

Note: Glycerin and glycerol are considered the same.

Abbreviations: CMC, carboxymethylcellulose; HP‐guar, hydroxypropyl‐guar; PEG, polyethylene glycol; TS‐polysaccharide, tamarind seed polysaccharide.

Osmoprotectants were added to 33 (~38%) of the artificial tears (Table 1). The different osmoprotectants in the Nordic artificial tears included L‐carnitine, erythritol, sorbitol, ectoine and trehalose. Electrolytes and buffers were found in 80 (~91%) tear substitutes, including a lot odifferent forms of electrolytes or buffers (Tables 2 and 3).

4. DISCUSSION

In the following section, the different preservatives, preservative‐free artificial tears and other constituents will be assessed, starting with the most well‐documented preservatives: quaternary ammonium compounds.

4.1. Preservative agents in artificial tears

4.1.1. Quaternary ammonium compounds

Quaternary ammonium compounds include BAK, CET and PQ‐1. These hydrosoluble compounds act by their detergent properties (Baudouin et al., 2010) and have been shown to enhance transcorneal penetration into the anterior chamber of the eye by interrupting the hydrophobic barrier of the corneal epithelium (Baudouin et al., 2010; Coroi et al., 2015; Goldstein et al., 2022). This, in turn, may result in a higher concentration of the drug localized at the target cell (Baudouin et al., 2010). A systematic review by Hedengran et al., however, showed no significant difference in terms of effectiveness between BAK and preservative‐free (PF) eye drops (Hedengran et al., 2020).

4.1.2. Benzalkonium chloride

Benzalkonium chloride is a quaternary ammonium used as a preservative in a wide range of artificial tears (Baudouin et al., 2010; Coroi et al., 2015; Figus et al., 2021; Goldstein et al., 2022; Kaur et al., 2009; Riedlova et al., 2023; Vitoux et al., 2020). It is typically found in concentrations between 0.005% and 2% (Coroi et al., 2015; Goldstein et al., 2022). The threshold concentration at which toxicity occurs is assessed to be 0.005% (Goldstein et al., 2022; Kolko et al., 2023). Clinical manifestations of BAK‐induced ocular surface toxicity include post‐administration discomfort such as foreign body sensation, burning sensation, ocular dryness and decreased tear break‐up time (TBUT) (Coroi et al., 2015; Goldstein et al., 2022). Benzalkonium chloride also causes goblet cell loss, which is related to reduced mucin expression and production (Coroi et al., 2015; Mantelli et al., 2011; Ribeiro et al., 2019). In addition, BAK has been shown to increase the cytoplasmic/nucleus ratio and lead to a reduced Schirmer's test value when comparing BAK‐containing eye drops to a control group (Baudouin et al., 2010).

The major consequence of hyperosmolarity of the tear film as seen in DED is apoptosis of corneal and conjunctival epithelial cells (Clouzeau et al., 2012; Labetoulle, Benitez‐Del‐Castillo, et al., 2022). In an in vitro study by Clouzeau et al., the cytotoxic effect of BAK alone and in combination with NaCl‐induced hyperosmolar conditions was investigated on Wong Kilbourne derivatives of Chang conjunctival epithelial cells (WKD‐cells). When a combination of both factors was present, cell death was significantly higher. This indicates that BAK toxicity is heightened in hyperosmotic conditions resembling the clinical condition observed in patients with DED. (Clouzeau et al., 2012). As a result, BAK contributes to the conditions that amplify its cytotoxic effect, creating an exacerbating vicious cycle (Baudouin et al., 2021; Kolko et al., 2023) (Figure 1).

FIGURE 1.

Benzalkonium chlorides (BAKs) self‐reinforcing adverse effects creating an exacerbating vicious cycle. TBUT, Tear break‐up time. The figure is created with Servier Medical Art.

Three different and unrelated studies examined the effect of BAK on either human corneal epithelial cells (HCE‐cells) (Vitoux et al., 2020), or WKD‐cells (Clouzeau et al., 2012; Datta et al., 2017). In all the studies, an elevation of reactive oxygen species (ROS), increased chromatin condensation and a production of apoptotic cell bodies were seen (Clouzeau et al., 2012; Datta et al., 2017; Vitoux et al., 2020). Datta et al. showed that BAK directly interferes with mitochondrial function by inhibiting mitochondrial O₂ consumption in human corneal epithelial primary cells at a 25‐ to 150‐fold lower concentration than in eye drops (Datta et al., 2017). One of the studies also showed a significant increase in extracellular ATP after BAK exposure to HCE‐cells, which can activate the ionotropic P2X receptors involved in the transmission of pain (Vitoux et al., 2020). Benzalkonium chloride has also been shown to induce a release of cytochrome C, subsequently leading to caspase activation. Moreover, BAK has been found to initiate caspase‐independent apoptosis (Buron et al., 2006; Clouzeau et al., 2012; Vitoux et al., 2020).

Interestingly, BAK is known to have a negative impact on human trabecular meshwork cells (Baudouin et al., 2010; Goldstein et al., 2022; Hamard et al., 2003). Hamard et al. found that BAK‐preserved eye drops led to a significant increase in all the apoptotic markers used (Apo2.7, annexin V binding and DNA content) (Hamard et al., 2003).

Although most research focusses on the direct toxic effect on the ocular surface, some studies have suggested that BAK might reduce corneal sensitivity. Martone et al. found that glaucoma patients treated with BAK‐preserved eye drops for at least 12 months had both reduced density of superficial epithelial cells as well as the number of nerves in the sub‐basal nervous plexus investigated by in vivo confocal microscopy (Martone et al., 2009). This might lead to reduced corneal sensitivity compared with untreated patients (Martone et al., 2009).

It has been reported that BAK exerts time‐ and dose‐dependent toxic effects (Baudouin et al., 2010; Coroi et al., 2015; Kolko et al., 2023). The detrimental effects have thus been shown to be cumulative and more severe with higher concentrations and repeated exposure (Debbasch et al., 2001; Kolko et al., 2023). In line with this, reducing the number of installations has been found to improve ocular tolerance (Baudouin et al., 2010).

4.1.3. Polyquaternium‐1

Polyquaternium‐1 is also a quaternary ammonium preservative compound (Coroi et al., 2015; Kathuria et al., 2021; Kaur et al., 2009; Martone et al., 2009; Walsh & Jones, 2019). The concentration of PQ‐1 in artificial tears is typically 0.001% (Walsh & Jones, 2019). Due to its high‐molecular weight (27 times greater than BAK) and lack of a hydrophobic region, it is not absorbed into mammalian cells and therefore is less likely to cause cytotoxic effects (Ammar et al., 2011; Coroi et al., 2015; Goldstein et al., 2022; Hedengran & Kolko, 2023; Kathuria et al., 2021; Walsh & Jones, 2019). Studies have shown less cytotoxic properties of PQ‐1 on the ocular surface compared with BAK (Coroi et al., 2015; Goldstein et al., 2022; Kaur et al., 2009; Walsh & Jones, 2019).

Two in vitro studies compared the effects of PQ‐1‐ and BAK‐containing eye drops in WKD‐cells. Both studies found better viability, less evidence of apoptosis and less production of ROS with PQ‐1‐containing solutions compared with BAK‐containing solutions (Brignole‐Baudouin, Riancho, Liang, & Baudouin, 2011; Brignole‐Baudouin, Riancho, Liang, Nakib, & Baudouin, 2011). Likewise, another in vitro study by Ammar et al. also showed a significantly higher percentage of viable corneal and conjunctival cells when treated with PQ‐1 preserved eye drops compared with BAK‐preserved eye drops (Ammar et al., 2011).

Furthermore, Whitson et al. demonstrated that PQ‐1‐containing solutions induced significantly less cytotoxicity than BAK (Whitson & Petroll, 2012). A study by Hedengran et al., examined the effect of PQ‐1‐preserved travoprost compared with BAK‐preserved in terms of viability, mucin and cytokine secretion on human conjunctival goblet cells. The lactate dehydrogenase assay showed no significant goblet cell loss when comparing PQ‐1‐preserved travoprost to the control, whereas BAK‐preserved travoprost caused a significant goblet cell loss. Immunohistochemical staining showed mucin secretion from goblet cells when incubated with PQ‐1 or BAK travoprost. Besides this, both PQ‐1 and BAK exhibited no significant increase in cytokine (interleukin‐6 and ‐8) secretion from the goblet cells compared with the control (Hedengran et al., 2022).

On the contrary, a study concluded that PQ‐1 evokes a degree of cytotoxicity and induces an even stronger inflammatory response in an NF‐kB‐dependent manner on HCE‐2 cells compared with BAK with the same concentration (Paimela et al., 2012). Another study noted a lower density of basal epithelial cells and a lower number of Langerhans cells when treating glaucoma patients with travoprost preserved with PQ‐1 compared with BAK (Marsovszky et al., 2014). It is, however, discussed whether these findings could be due to the lack of detrimental effects from BAK in this study (Marsovszky et al., 2014).

4.1.4. Cetrimide

Cetrimide is—such as BAK and PQ‐1—a quaternary ammonium compound. The adverse effects caused by CET are closely related to the adverse effects caused by BAK. Debbash et al. tested and compared different quaternary ammonium compounds (CET, benzododecinium bromide and three different kinds of BAK with different hydrocarbon chain lengths) to five other preservatives (Debbasch et al., 2001). The various preservatives were tested on WKD‐cells at different concentrations, including the concentration used in most eye drops being 0.01% (Debbasch et al., 2001). They showed that the quaternary ammoniums tested were the most cytotoxic preservatives as it activated apoptosis, increased ROS production and disrupted cell membrane integrity. Even after 24 h of cell recovery, significant cellular damage and chromatin condensation were still found in the cells treated with the above‐mentioned quaternary ammonium compounds with a concentration of 0.001% and above (Debbasch et al., 2001).

4.1.5. Alternative preservatives

Besides quaternary ammonium compounds, other preservatives are available, here starting with the group: stabilized oxychloro complex.

4.1.6. Stabilized oxychloro complex

Stabilized oxychloro complexes are chemical oxidizing agents, including Purite®, OcuPure® and Oxyd®. These contain 99.5% chlorite, 0.5% chlorate and trace amounts of chlorine dioxide (Goldstein et al., 2022; Hedengran & Kolko, 2023; Kathuria et al., 2021; Kaur et al., 2009; Walsh & Jones, 2019). When applied topically, the preservative vanishes and is converted to sodium, oxygen, water and chloride ions. Although SOC is a chemical oxidant, there is no in vivo or in vitro evidence of mutagenicity or carcinogenicity (Kaur et al., 2009). In a solution, SOC produces chloride dioxide‐free radicals, which provide the oxidizing action resulting in bactericidal, fungicidal and virucidal activity (Goldstein et al., 2022; Hedengran & Kolko, 2023; Kathuria et al., 2021; Kaur et al., 2009; Tu, 2014; Walsh & Jones, 2019). Oxidative preservatives such as SOC are usually small molecules, which can penetrate cell membranes and interfere with cellular function (Kaur et al., 2009). They can destabilize cell membranes, but to a lesser degree than detergent preservatives such as BAK. In contrast to microorganisms, mammalian cells are equipped with antioxidants in the tear film such as catalases. The toxic effect on eukaryotic cells is therefore anticipated to be negligible (Kaur et al., 2009). One study compared the effect of ophthalmic solutions containing BAK, PQ‐1, disodium edetate and Purite® using HCE‐cells. Purite® showed significantly less cell death compared with BAK, but when compared to the other tested preservatives, Purite® reduced the cell viability the most after 1 h of exposure (Xu et al., 2013).

OcuPure® is closely related to Purite®. When installed in the eye and afterwards exposed to light, it is converted to sodium, oxygen, water and chloride ions (Walsh & Jones, 2019). The evidence outlined above suggests a less cytotoxic effect by OcuPure® compared with BAK, but still some toxicity compared with preservative‐free eye drops. However, there is not enough knowledge on the effect of this preservative on the human eye.

Oxyd®, as mentioned above, is also a vanishing preservative. Maloni et al., conducted an in vitro experiment, applying various preservatives to HCE‐cells over three time intervals, considering recovery and assessing cumulative effects (Meloni et al., 2010). They found that cell viability decreased to 71% after 24 h of treatment with 24 hours of post‐incubation. Oxyd® treatment for 72 h resulted in a significant decrease to 4.5%. Additionally, Oxyd® treatment led to the release of chemokines, including IL‐8. Furthermore, immunofluorescence analysis showed a thinned and damaged epithelium with an abnormal occludin distribution after treatment with Oxyd® (Meloni et al., 2010).

4.1.7. Polyhexamethylene biguanide

Polyhexamethylene biguanide is a broad‐spectrum antiseptic compound and is a mixture of polymeric biguanides. The biguanide residues are strong bases, which means they are highly positively charged at a physiological pH value. It works with a broad‐spectrum of action even at low concentrations and has been shown to be better tolerated than more widely used preservatives such as BAK (Niro et al., 2022). Intriguingly, PHMB does not readily penetrate the cornea to the anterior chamber, which may be due to an electrostatic interaction between negatively charged corneal proteoglycans and the positively charged biguanides (Niro et al., 2022).

Most of the existing research according to PHMB is based on its use in contact lens solutions, its antiseptic involvement in surgery and as a treatment for acanthamoeba keratitis. Studies concerning PHMB in artificial tears have not been conducted to our understanding.

A randomized, double‐masked, placebo‐controlled phase I study evaluated the safety and tolerability of PHMB 0.04%, 0.06% and 0.08% on healthy volunteers for 2 weeks (Papa et al., 2022). The study aimed to recruit 90 subjects with 27 in each PHMB group and nine in the placebo group. The rates of dose‐limiting adverse events (DLAEs) leading to interruption of dosing, mild adverse events (not dose‐limiting) and incidental adverse events (unrelated to treatment) were compared. The study showed that 5/90 subjects developed DLAE located on the ocular surface of which: two were using PHMB 0.06% and three were using PHMB 0.08%. These DLAEs developed within <1–4 days. There were no significant differences in DLAE between treatment groups. There were statistically significant differences in corneal stain scores of different PHMB concentrations with a higher corneal staining score for PHMB 0.08% Corneal staining had returned to baseline in almost all subjects by Day 21 (1 week after discontinuation of treatment). The findings suggest that PHMB 0.04–0.06% is safe and that higher PHMB concentrations may be safe, but the long‐term adverse effects of PHMB are still unknown (Papa et al., 2022).

4.1.8. Phenoxyethanol

Phenoxyethanol is a preservative commonly used in cosmetics. Phenoxyethanol increases the permeability of the ion channels localized on the cell membrane, essentially inhibiting the synthesis of microbial genes (Kim et al., 2023). Phenoxyethanol is allowed in consumer products in a concentration of up to 1.0% (Kim et al., 2023; Wang et al., 2020). However, another study made by Wang et al. showed toxic effects on immortalized human meibomian gland epithelial cells (IHMGECs) at half the concentration of the upper recommended limit (i.e. 0.5%) (Wang et al., 2020). At this level, it significantly reduced Akt‐signalling in IHMGECs after 30 min of exposure. When the concentration was further decreased to one‐tenth, it significantly decreased cell survival following an exposure of 24 h (Wang et al., 2020).

A study compared 2‐phenoxyethanol, BAK and triton X‐100 on an in vitro mini‐cornea model with immortalized human corneal epithelial cells and keratocytes. These cell cultures were used for an irritation test (Kim et al., 2023). The results of phenoxyethanol treatment showed no significant difference between the concentrations of 0 to 1%, but when treated with a concentration of 2%, the cell viability reduced to 27.59 ± 2.72% (Kim et al., 2023). A haematoxylin and eosin staining of the mini‐cornea model treated with 2% phenoxyethanol, was also conducted. The cells showed an irregular surface and the cells dissociated from each other compared with the control (Kim et al., 2023). See Table 4 for a summary and overview of the adverse effects caused by preservatives in artificial tears.

TABLE 4.

Summary and overview of the adverse effects caused by preservatives in artificial tears.

Preservative	Adverse effects
BAK	Discomfort after administration, including foreign body sensation, burning sensation and ocular dryness BAK induces apoptosis of corneal and conjunctival epithelial cells BAK causes a significant expression of the apoptosis marker Apo2.7 Goblet cell loss and reduced mucin expression ↓ TBUT ↓ Schirmer's test value ↓ corneal sensitivity BAK toxicity is enhanced in hyperosmotic conditions ↑ ROS ↑ extracellular ATP activating ionotropic P2X receptors involved in the transmission of pain ↑ cytotoxicity, chromatin condensation and production of apoptotic cell bodies BAK might inhibit mitochondrial O2 consumption BAK has a negative impact on human trabecular meshwork cells
PQ‐1	Less likely to cause cytotoxic effects as it is not absorbed into mammalian cells Better cell viability, less apoptosis, and less production of ROS when compared with BAK
CET	The adverse effects caused by CET are closely related to the side effects caused by BAK ↑ apoptosis, cellular damage and chromatin condensation ↑ ROS Disruption of the cell membrane integrity
SOC	Only a few published studies exist on the effect of these preservatives on the human eye Purite®: Less cell death compared with BAK OcuPure®: Studies concerning OcuPure® have not been conducted Oxyd®: Time‐dependent decrease in cell viability It damages the epithelium leading to an abnormal occludin distribution Release of chemokines, including IL‐8
PHMB	Studies concerning PHMB in artificial tears have not been conducted
Phenoxyethanol	Concentration‐dependent decrease in cell viability Concentration‐dependent reduction in Akt‐signalling

Abbreviations: BAK, benzalkonium chloride; CET, cetrimide; PHMB, polyhexamethylene biguanide; PQ‐1, polyquaternium‐1; ROS, reactive oxygen species; SOC, stabilized oxychloro complexes; TBUT, tear break‐up time.

4.2. Preservative‐free artificial tears

Preservative‐free (PF) artificial tears can be packaged in single‐dose units with a twist‐off cap, typically containing between 0.1 and 1 mL of fluid, or in multi‐dose units with a filter system. A clinical multicentre cross‐sectional epidemiologic survey study by Jaenen et al. investigated the subjective effect on patients switching from preserved to PF eye drops (Jaenen et al., 2007). Both groups included almost the same order of symptoms. However, the prevalence of discomfort during installation, stinging or burning sensation, foreign body and dry eye feeling, eyelid itching and tearing was significantly lower in the group receiving PF eye drops (Jaenen et al., 2007). An in vitro study compared BAK‐preserved eye drops and PF eye drops on HCE‐cells (Xu et al., 2013). PF formulations were found to exhibit significantly lower cytotoxicity to cultured corneal epithelial cells than their preservative‐containing counterparts (Xu et al., 2013). PF artificial tears are known to be less irritating to the ocular surface (Figus et al., 2021; Hakim & Farooq, 2022).

Besides preservatives, other different constituents are added to artificial tears. In the following section, lubricants, osmoprotectants, lipids and other constituents will be assessed according to their efficacy, safety and toxicity.

4.3. Other constituents in artificial tears

4.3.1. Lubricating agents

The lubricating agents found in artificial tears in the Nordic market are carbomer, carboxymethylcellulose (CMC), dextran, glycerin (glycerol), hypromellose, hyaluronate, hyaluronic acid (HA), hydroxypropyl‐guar (HP‐guar), polyethylene glycol (PEG), polyvinyl alcohol, polysorbate, povidone, propylene glycol, sodium hyaluronate (SH), sorbitan monooleate, sorbitan oleate and tamaring seed polysaccharide (TS‐ polysaccharide).

Hyaluronate, hyaluronan/hyaluronic acid (HA) and sodium hyaluronate (SH) are all related forms of the same molecule (Becker, Bergfeld et al., 2009). Sodium hyaluronate is the sodium salt of hyaluronic acid, and when dissolved in water all these chemicals are chemically identical.

4.3.2. Carbomer

Carbomer resins are synthetic, high‐molecular‐weight, nonlinear polymers of predominantly acrylic acid, cross‐linked with a polyalkenyl polyether (Johnson et al., 2008).

A randomized, double‐masked study compared the effectiveness of artificial tears containing 0.3% carbomer 934 or 0.18% SH in treating moderate dry eye (Johnson et al., 2008). A total of 65 subjects were included and randomized to receive either artificial tears containing carbomer 934 or SH. After a month of treatment, they found that both treatments reduced the severity of symptoms equally. Sodium hyaluronate outperformed carbomer in improving the integrity of the ocular surface. However, neither had a lasting effect on improving TBUT and ocular staining (Johnson et al., 2008).

An in vitro study using WKD‐cells investigated cytotoxic effects of carbomer 934P and HA with and without preservative (BAK), BAK alone was also investigated (Debbasch et al., 2002). Furthermore, their interactions with BAK were studied to understand how these interactions might reduce the toxic effects of BAK. Their usual commercial concentrations were used to treat WKD‐cells for either 15 min or 15 min with 24 h of cell recovery, respectively. After 15 min of treatment, no significant differences were found between unpreserved formulations, whereas preserved HA was found to be significantly less cytotoxic than preserved carbomer (p < 0.0001). After 24 hours of cell recovery, no alteration was found with unpreserved and preserved HA, whereas a significant decrease in cell viability was observed with unpreserved and preserved carbomer 934P. This cytotoxicity, however, was significantly less than that observed with BAK alone, although the same concentrations of preservative were used. Regarding ROS, the four formulations decreased the production, whereas BAK alone significantly increased the production (Debbasch et al., 2002).

4.3.3. Carboxymethylcellulose

Carboxymethylcellulose, a high‐molecular‐weight polysaccharide, is utilized for its ability to enhance ocular surface retention due to its viscous consistency and mucoadhesive attributes (Lee et al., 2015). The safety and efficacy of CMC 1% in improving tear film stability post‐phacoemulsification was evaluated by Yao et al., This prospective, multicentre, open‐label, randomized, controlled study demonstrated that treatment with CMC 1% is safe and improves tear film stability following phacoemulsification for age‐related cataract (Yao et al., 2015).

Two separate studies compared the effectiveness of CMC 0.5% and SH in treating dry eye symptoms. The first study, involving 67 patients with mild‐to‐moderate DED, found equivalent efficacy in managing dry eye symptoms between the two eye drops (Lee et al., 2011). The other study, a prospective randomized case series, yielded similar results, demonstrating that both treatments were effective and well‐tolerated in reducing symptoms of DED post‐surgery (Mencucci et al., 2015).

Furthermore, a comparative study examined the cytotoxicity and wound‐healing effect of CMC and HA on HCE‐cells (Lee et al., 2015). The findings suggested that both substances, when used in artificial tear formulations promoted corneal wound healing and were safe without any significant toxic effects (Lee et al., 2015).

4.3.4. Hyaluronic acid

Hyaluronic acid/hyaluronan is a member of the large family of glycosaminoglycans, which are the main components of the extracellular matrix (Muller‐Lierheim, 2020). Hyaluronic acid possesses excellent gelling properties due to its rapid water‐binding capability. It serves as a natural component of tears synthesized by epithelial cells. Compared with other tear substitute formulations, HA has been shown to enhance tear film viscosity and stability. Eye drops typically contain HA concentrations ranging from 0.18% to 0.2%. Its mucoadhesive properties and excellent gelling properties contribute to greater relief of dry eye symptoms allowing corneal wettability (Aragona et al., 2002; Ballesteros‐Sanchez et al., 2023; Pinto‐Fraga et al., 2017).

A study showed a significant relief of DED symptoms and improved rose bengal staining in patients treated with 0.1% and 0.18% HA for 1 month (Muller‐Lierheim, 2020). Moreover, they showed that long‐term use of a 0.15% SH‐containing artificial tear reduced ocular damage in DED patients (Muller‐Lierheim, 2020).

In a clinical trial by Pinto‐Fraga et al., the efficacy and safety of an artificial tear containing 0.2% HA were evaluated against a 0.9% saline solution in subjects with DED (Pinto‐Fraga et al., 2017). Approximately 31.2% of the subjects using eye drops containing HA reported five points or more improvement in ocular surface disease index (OSDI) scores compared with the saline solution, with no significant adverse effects observed (Pinto‐Fraga et al., 2017).

An in vitro study by Rangarajan R et al., compared HA/HP‐guar dual‐polymer formulations with single‐polymer formulations containing either HA or HP‐guar alone on immortalized human corneal epithelial cells (Rangarajan et al., 2015). HA/HP‐guar exhibited superior cell viability, surface retention and desiccation protection compared with HP‐guar or HA alone (p < 0.001) (Rangarajan et al., 2015).

Mucoadhesion, crucial for the effectiveness of eye drops, is related to ocular residence time and therefore to the effectiveness of eye drops. The properties of hyaluronan in eye drops depend on chain length and concentration. Hyaluronic acid with high‐molecular weight has a higher water‐binding capacity leading to a longer retention time on the ocular surface (Muller‐Lierheim, 2020). A prospective, multicentre, open‐label study by Medic et al. randomized 47 patients to compare the effects of high‐molecular‐weight HA with comparator eye drops, containing low‐molecular‐weight HA (Medic et al., 2023). The high‐molecular‐weight HA group exhibited a notable decrease in drop‐application frequency and OSDI scores compared with the low‐molecular‐weight HA group (Medic et al., 2023).

Another study evaluated the binding affinity between HA and MUC2 by rheological analysis, fluorescence analysis and surface plasmon resonance analysis (Guarise et al., 2023). They showed that the mucoadhesive performance of linear, natural HA was correlated with the molecular mass (MM) and exponentially related to the concentration of HA. The results showed a linear correlation between the MM of HA and the binding affinity with MUC2. Other emollients (PEG 400, CMC, glycerol) and gelling agents (guar gum) did not show the same mucoadhesive properties, except for xanthan gum. Even after simulating the pathological condition of DED by decreasing the MUC2 or oleic acid concentration, the mucoadhesive performance of high MM HA was still confirmed (Guarise et al., 2023).

Lastly, a study examined the cytoprotective effects of HA and hypromellose against DNA damage induced by thimerosal in conjunctival cells (Ye et al., 2012). Cells pretreated with HA and hypromellose exhibited increased cell survival, reduced DNA damage and diminished ROS production compared with cells exposed to thimerosal alone (Ye et al., 2012).

4.3.5. Hydroxypropyl‐guar

Hydroxypropyl‐guar, a viscous mucomimetic polymer, has been shown to reduce surface friction and improve DED symptoms (Wang et al., 2010). It works by interacting with lipid components of the tear film decreasing tear evaporation and may increase tear film stability (Jacobi et al., 2012). A prospective, randomized, clinical, single‐centre study compared the safety and effect of PF eye drops with HP‐guar with Tamarindus indica seed polysaccharide eye drops (Jacobi et al., 2012). HP‐guar seems to be slightly more effective in protecting the ocular surface by decreasing tear film evaporation. HP‐guar becomes more viscous with increased cross‐linking, and the drier the ocular surface is. This results in a strong gel‐like structure (Jacobi et al., 2012). Another randomized, double‐masked crossover study demonstrated that HP‐guar‐containing artificial tears significantly improved TBUT in patients with DED for up to 60 min compared with eye drops without HP‐guar (Ousler et al., 2007). This suggests a long‐lasting protection from desiccation (Ousler et al., 2007).

4.3.6. Polyethylene glycol

A study evaluated the efficacy and safety of a test product of polyethylene glycol (PEG) 400/propylene glycol and HP‐guar in comparison with Optive® (containing CMC and glycerin) (Davitt et al., 2010). They found that patients treated with the test product exhibited significantly lower (mean) corneal and conjunctival staining scores at Days 14 and 42. Patients in both groups reported significant reductions in scores for ocular dryness, sandy/gritty and burning sensation. Likewise, the OSDI score was significantly lower for both groups at Day 42 compared with baseline (Davitt et al., 2010).

4.3.7. Propylene glycol

A phase IV, multicentre, open‐label and interventional trial evaluated symptom relief in patients with DED following a single drop of propylene glycol (PG) and HP‐guar (Systane® Complete) (Silverstein et al., 2020). Patients instilled one drop in each eye and changes from baseline, soothing sensation and tolerability scores were assessed at zero, four‐ and 8 h post dose. They found that the artificial tear provided immediate and sustained symptom relief for 8 hours and was well‐tolerated (Silverstein et al., 2020). Besides this, a randomized double‐masked study of dry eye patients with noninflammatory meibomian gland disease or aqueous tear deficiency showed that artificial tears containing the demulcents PG and PEG resulted in longer precorneal residence time compared with saline alone (Paugh et al., 2008).

The combined effect of PEG‐PG on mucin secretion within a hyperosmotic stress–based explant model was analysed (Panigrahi et al., 2023). Human corneoscleral rims were obtained from a healthy donor post‐corneal keratoplasty and hyperosmolar stress mimicking DED was induced. Gene expression and mucin secretion were analysed. In hyperosmolar conditions, the corneoscleral rim upregulated NFAT5 (a marker for increased osmolarity) and the expression of MUC5AC and MUC16 was reduced, whereas they showed induction of MUC5AC and MUC16 upon treatment with PEG‐PG (Panigrahi et al., 2023).

A 90‐day, prospective, randomized, observer‐masked parallel‐group study by Labetoulle et al. aimed to demonstrate non‐inferiority of HP‐guar/PEG/PG (Systane® Ultra) compared with CMC (Optive®) with the primary efficacy endpoint being total ocular surface staining (OSFS) score change from baseline (Labetoulle et al., 2017). Safety was also evaluated. Patients included were diagnosed with dry eye before screening and were using BAK‐free eye drops at least once per day for ≥3 months. The treatments decreased OSFS, alleviated objective signs and patient‐reported symptoms of dry eye and they scored similarly for treatment effectiveness and convenience. They concluded that HP‐guar/PEG/PG was non‐inferior to CMC (Optive®) (Labetoulle et al., 2017).

4.3.8. Sodium hyaluronate

Sodium hyaluronate (SH) is a glycosaminoglycan disaccharide with a huge capacity to bind water, 1000‐fold its weight (Johnson et al., 2008). It has lubricant and viscoelastic properties and contributes to create mechanical protection for cells (Fariselli et al., 2018). A randomized, double‐blinded study by Aragona et al. investigated the effect of SH on the ocular surface of patients suffering from dry eye (Aragona et al., 2002). The study randomly assigned patients to receive either artificial tears containing SH 0.15% in phosphate buffered saline or physiological sodium chloride 0.9% for 3 months. Evaluation of impression cytology, slit lamp examinations and subjective symptoms were conducted at one, two and 3 months. After 3months of treatment, it was observed that long‐term use of SH‐containing artificial tears reduced ocular surface damage in patients with dry eye (Aragona et al., 2002).

A meta‐analysis compared randomized controlled trials regarding the use of SH combined with conventional treatment after cataract surgery (Wen et al., 2020). After 1 month of treatment, the combination group exhibited a significantly higher total effective rate compared with the control group. Additionally, dry eye symptom score, TBUT, OSFS score and Schirmer I test results were all significantly better in the combination group compared with the control group (Wen et al., 2020).

A multicentre, observer‐masked, randomized non‐inferiority study compared SH alone and SH + PEG800 (Labetoulle, Mortemousque and Group, 2022). The performance and safety of the two artificial tears on patients with moderate‐to‐severe DED were assessed. Both interventions demonstrated significant improvement in signs and symptoms of DED. They found that the safety profiles were satisfactory in both groups and that the artificial tears were non‐inferior to each other in terms of change in OSFS score between baseline and Day 28. However, the OSFS scores were significantly greater in the SH + PEG 8000 group compared with the SH group on Day 90 suggesting that SH + PEG 8000 might grant long‐term benefits compared with SH alone (Labetoulle, Mortemousque and Group, 2022).

The ability of SH, CMC and hypromellose to enhance water retention and to protect HCE‐cells from dehydration were compared in an in vitro study (Zheng et al., 2013). The lubricants were tested at 0.1% and 0.3% concentrations. They found that the weight of filter papers soaked in SH at both concentrations was significantly heavier than that of those soaked in CMC and hypromellose at almost all time points evaluated. This indicated less evaporation of water with the SH‐solution. Also, the survival rate of HCE‐cells was significantly greater when pretreated with SH than CMC and hypromellose following dehydration. Overall, they found SH to be superior to the other two lubricants regarding the prevention of water retention and protection of HCE‐cells against dehydration (Zheng et al., 2013).

A study by Duan et al., compared the clinical efficacy of SH (group A), polyethylene glycol (group B) and dextran‐70 (group C) artificial tears in relieving dry eye after cataract surgery (Duan & Tang, 2021). Ninety‐nine patients were divided into groups, respectively. TBUT, Schirmer I test, OSFS, dry eye symptom score and incidence of ocular irritation were assessed. They found that TBUT, Schirmer I test, OSFS and dry eye symptom score were significantly (p < 0.05) better in group A and B compared with group C. No significant differences were seen between group A and B. The incidence of ocular irritation symptoms was significantly lower in group A and B than that in group C, and the incidence in group A was significantly lower compared with group B (Duan & Tang, 2021).

4.3.9. Polysorbate 80

Polysorbate 80 is an amphipathic, nonionic surfactant composed of fatty acid esters of polyoxyethylene sorbitan (Kerwin, 2008) It is an emulsifying agent with surfactant properties, important for stabilizing the oil and water phases in artificial tears (Dhahir et al., 2021). A study assessed acute cytotoxicity on human corneal epithelial cells (HCE‐T) during clinically relevant exposure times. BAK‐preserved ophthalmic formulations exhibited concentration‐ and time‐dependent cytotoxicity, while nonionic excipients as polysorbate 80 showed no acute cytotoxicity. Cell viability was assessed through resazurin reduction to highly fluorescent resorufin, and the expression of the tight junction proteins in HCE‐T cells by immunofluorescence staining (Hakkarainen et al., 2016).

4.3.10. Tamarind seed polysaccharide

Tamarind seed polysaccharide (TSP/TS‐polysaccharide) is derived from the tamarind seed. The configuration of TSP gives a mucin‐like structure resembling MUC1, making it a promising treatment for dry eye syndrome (Rolando & Valente, 2007). An open‐label randomized clinical study compared TSP 0.5%, TSP 1% and HA 0.2% over 90 days in 30 patients. Both TSP formulations matched HA in tolerability and efficacy, assessed via symptom scoring, tear film stability and corneal staining. Notably, TSP 1% improved subjective symptoms, including blinking difficulty, burning and foreign body sensation, highlighting its potential as an alternative to HA (Rolando & Valente, 2007).

A multicentre, randomized, double‐masked study evaluated the efficacy of a combination of TS‐polysaccharide 0.2% and hyaluronic acid (HA) 0.2% (Xiloial) versus carmellose sodium (Optive) in 48 patients with moderate dry eye disease (DED). Patients treated with HA and TS‐polysaccharide showed a statistically significant improvement in the OSDI score at the end of the study compared with the baseline and control groups. The combination of HA and TS‐polysaccharide were equally effective as carmellose sodium in reducing BUT and the extent of injury assessed by corneal and conjunctival staining. Non‐significant changes were recorded for Schirmer I test (Barabino et al., 2014).

4.3.11. Dextran 70, glycerol, hypromellose, PEG 300, polyvinyl alcohol, and povidone

One study compared dextran‐70 with SH and PEG and is mentioned in the SH section (Duan & Tang, 2021). Hypromellose is compared with SH and CMC in an in vitro study and is also mentioned in the sodium hyaluronate section (Zheng et al., 2013).

Despite their widespread use, no articles have been found on the adverse effects, efficacy and safety of glycerol, PEG 300, polyvinyl alcohol, sorbitan monooleate, sorbitan oleate, or povidone. Understanding these aspects is crucial for consumer safety, highlighting a need for further research in this area.

4.4. Osmoprotectants

The osmoprotectants found in artificial tears in the Nordic market are ectoine, erythritol, levocarnitine (L‐carnitine), sorbitol and trehalose.

4.4.1. Ectoine

Ectoines, organic solutes produced by bacteria facing extreme stress, have shown effects on lipid layers. A recent study by Dwivedi et al. investigated the impact of ectoine on artificial lipid films mimicking tear fluid (Dwivedi et al., 2014). The study utilized various biophysical techniques such as surface activity analysis and topology analysis, to assess changes induced by ectoine. Surface activity studies revealed that ectoine increased the spacing between lipid headgroups and weakened interactions within the lipid layer. Additionally, topology analysis showed the formation of droplet‐like structures in the presence of ectoine, particularly when nonpolar components were present alongside phospholipids and cholesterol esters (Dwivedi et al., 2014). These findings mirror observations made in studies on meibomian lipids, suggesting that ectoine induces similar structural alterations in both artificial and natural lipid layers. The increased fluidity induced by ectoine could potentially alleviate symptoms of DED (Dwivedi et al., 2014).

4.4.2. L‐carnitine

L‐carnitine is a naturally occurring amino acid best known for its role in the mitochondrial oxidation of long‐chain fatty acids. A study by Khandekar N. et al. evaluated the osmoprotective properties of L‐carnitine on human corneal limbal epithelial cell volume and apoptosis in hyperosmolar conditions (Khandekar et al., 2013). In hyperosmotic medium (500 mOsm), the number of shrunken cells and damaged/dead cells was six‐ and threefold times higher compared with isotonic conditions (300 mOsm). Moreover, tumour necrosis factor‐α (TNF‐α) concentration and caspase ‐3, ‐7, ‐8 and ‐9 activities were all significantly elevated in hyperosmolar conditions compared with isotonic. The addition of L‐carnitine to hyperosmolar conditions significantly reduced the concentration of TNF‐α and caspase‐9 activity. Furthermore, the cell volume was partly restored and the percentage of damaged/dead cells was reduced to levels observed under isotonic conditions (Khandekar et al., 2013).

Another study investigated the effects of osmoprotective agents (L‐carnitine, Betaine, Taurine) and mucoadhesive polymers (HA and hypromellose) on immortalized HCE‐cells in hyperosmolar conditions (Lopez‐Cano et al., 2021). They found that the osmoprotective agents and mucoadhesive polymers protect corneal epithelial cells from cell death under chronic hyperosmolarity conditions. Pre‐incubation with betaine (150 mM and 200 mM) provided the highest cell survival against hyperosmolarity whereas HA 0.4% was the most effective mucoadhesive polymer (Lopez‐Cano et al., 2021).

The osmoprotective effects of L‐carnitine were also shown in a study by Corrales et al. (2008). They assessed, among others, the effect of osmoprotectants including L‐carnitine and erythritol on the activation state of mitogen‐activated protein (MAP) kinases in HCE‐cells in both physiological and hyperosmolar conditions (Corrales et al., 2008). They found that L‐carnitine and erythritol alone significantly lowered levels of activated MAP‐kinases in response to hyperosmolar stress. Furthermore, they lowered ratios of phosphorylated to total kinases in isotonic media (Corrales et al., 2008).

Even though carnitine is not recognized as a normal constituent of tears in healthy subjects, a study by Pescosolido N et al., investigated the levels of L‐carnitine and short‐chain esters in tears from patients with dry eyes (Pescosolido et al., 2009). Tears were collected from 10 healthy subjects and 10 patients with dry eyes. Carnitine levels were assessed by high‐performance liquid chromatography‐mass spectrometry. Carnitine levels were significantly lower in tears from individuals with dry eyes compared with healthy controls. They suggest that the damage to ocular surface cells may be partially due to an imbalance in the concentration of carnitine molecules in the tear film (Pescosolido et al., 2009).

4.4.3. Trehalose

Trehalose is a natural disaccharide consisting of two glucose molecules. It is known to stabilize proteins and membranes, inhibit oxidative stress and prevent cells from denaturation (Ballesteros‐Sanchez et al., 2023). It also has water‐retaining properties (Fariselli et al., 2018; Pinto‐Bonilla et al., 2015).

An open‐label, pilot study assessed different parameters in conjunctival epithelium before and after treatment with trehalose/hyaluronate tear substitutes in 15 DED patients (Fariselli et al., 2018). Patients treated with trehalose/hyaluronate tear substitute had a significant improvement of OSDI scores, corneal staining, conjunctival staining and impression cytology compared with baseline. A decrease in ocular discomfort symptoms and surface damage was especially seen after 2 months (Fariselli et al., 2018).

Trehalose‐based artificial tears, such as Thealoz® Duo, have shown superior effects compared with Systane®. A randomized, single‐centre, open‐label, crossover study randomized 17 patients with moderate‐to‐severe DED to treatment with either Thealoz® Duo or Systane® (Pinto‐Bonilla et al., 2015). Satisfaction increased with Thealoz® Duo treatment, showing statistically significant advantages over Systane® in two parameters: symptoms of DED and their impact on work. No statistically significant advantages for Systane® were observed over Thealoz® Duo in any measured parameter and no adverse events were reported (Pinto‐Bonilla et al., 2015).

The efficacy and safety of trehalose in the treatment of DED was the topic of a systematic review by Ballesteros (Ballesteros‐Sanchez et al., 2023). A total of 10 RCTs were included. And parameters such as OSDI, TBUT, tear film thickness, tear meniscus height, Schirmer I test, OSFS and patient satisfaction were assessed. They found that trehalose tear substitute treatments achieved higher improvement than the control group interventions in all reported variables. Also, no adverse events were reported after instillation of trehalose‐tears substitutes (Ballesteros‐Sanchez et al., 2023).

While trehalose‐based artificial tears are considered safe, a study has attempted to show that trehalose has a cytoprotective ability (Hill‐Bator et al., 2014). A study by Hill‐Bator A et al., investigated the cytoprotective ability of Thealoz®, a trehalose‐based artificial tear, compared with other artificial tears during desiccation of HCE‐cells for 5, 15, 30 and 45 min (Hill‐Bator et al., 2014). Thealoz® showed the highest effectiveness in preventing cell death from desiccation and also in keeping normal cellular membrane function (Hill‐Bator et al., 2014).

Two other studies evaluated the protective effects of trehalose. The first study evaluated the effectiveness of trehalose (Thealoz®) in protecting reconstructed human corneal epithelium from desiccation (Hovakimyan et al., 2012). The cells were divided into a control group (I), a desiccation group without pretreatment with trehalose (II), and with pretreatment (III). The protective role of trehalose was evaluated using in vitro cell viability assays. They found that 32% of the cells survived desiccation without pre‐incubation against 98% in trehalose pre‐incubated tissue (Hovakimyan et al., 2012). The second study by Hill‐Bator A et al. examined the effects of trehalose‐based eye drops on HCE‐cells under desiccation conditions (Hill‐Bator et al., 2014). When cultured HCE‐cells were subjected to desiccation stress for 5 to 45 min, trehalose‐containing artificial tears, unlike many other artificial tears, effectively preserved the viability and functionality of corneal epithelial cells during desiccation stress. This was evidenced by maintained cell viability, reduced cellular damage, and preserved cellular function compared with untreated cells under similar conditions. Moreover, trehalose maintained the integrity of the cells (Hill‐Bator et al., 2014).

Trehalose is shown to be more effective on some parameters than commonly used SH. This is shown in the following study (Caretti et al., 2019). The study compared carbomer sodium hyaluronate trehalose (trehalose group) with SH tears, on subjects after phacoemulsification. Objectives such as TBUT, corneal and conjunctival staining and subjective OSDI scores were assessed. They showed that the trehalose group was more effective on some parameters (TBUT, OSDI) than SH (Caretti et al., 2019).

Furthermore, trehalose has also been investigated in regards of its role in promoting corneal epithelial healing following corneal cross‐linking (CXL) treatment for keratoconus (Ozek & Kemer, 2018). The first operated eye of each patient was treated with SH alone and the second operated eye was treated with trehalose‐SH. The patients were followed up every day until complete re‐epithelialization was observed. Results from the study indicated that the application of trehalose promoted faster and more efficient healing of the corneal epithelium after CXL compared with the control group treated with SH alone (Ozek & Kemer, 2018). This was evidenced by a reduced size of epithelial defects and faster closure rates in the trehalose‐treated group. Moreover, histological analysis revealed enhanced epithelial cell migration and proliferation in the trehalose‐treated corneas, indicating an improved overall healing response (Ozek & Kemer, 2018).

4.4.4. Erythritol and sorbitol

Erythritol is mentioned in one study, together with L‐carnitine, where it was shown to have osmoprotective effects against the activation of MAP‐kinases in HCE‐cells in hyperosmolar conditions (Corrales et al., 2008). The study is mentioned in the L‐carnitine section.

Even though Sorbitol is used in 21 artificial tears in the Nordic market, no articles have been conducted regarding this osmoprotectant isolated. To confirm safety, understanding these aspects is crucial, highlighting a need for further research in this area.

4.5. Lipids

Lipids found in the Nordic market are castor oil, cetalkonium chloride, lanolin oil, liposomes, mineral oil, oleic acid, paraffin oil, perilla seed extract, petroleum jelly/vaseline, phosphatidylglycerol, phospholipid, sacha inchi seed oil, sea buckthorn oil, sea buckthorn seed oil, sorbitan tristearate, soybean oil, soy lecithin and triglycerides.

4.5.1. Castor oil

Castor oil is a vegetable oil extracted from the seeds of the castor bean plant (Ricinus communis).

A randomized, double‐masked, placebo‐controlled crossover clinical trial tested homogenized castor oil artificial tears for the treatment of patients with noninflamed obstructive meibomian gland dysfunction (MGD) (Goto et al., 2002). Twenty patients with noninflamed MGD whose symptoms had not improved sufficiently despite conventional treatment (lid hygiene and/or topical therapy with artificial tears, corticosteroids or systemic antibiotics) were selected and examined. Symptoms were assessed as well as tear interference grade, tear evaporation, TBUT, OSFS, rose bengal scores and meibomian gland orifice obstruction. Every score and test showed statistically significant improvement, besides OSFS (p = 0.06) compared with placebo. Furthermore, the artificial tear was well‐tolerated, and none of the subjects reported irritation or severe blurring (Goto et al., 2002).

Similar results were seen in a randomized parallel, longitudinal and investigator‐masked study. The study determined the effect of a castor oil emulsion compared with a conventional aqueous artificial tear (Khanal et al., 2007). A total of 53 patients with mild‐to‐moderate dry eye were randomized into two groups. Tear production, evaporation, lipid layer structure and osmolality were measured before and 30 days after the use of the drops. Evaporation was measured with a modified Servo‐Med EP‐1 Evaporimeter (Stockholm, Sweden). Changes in evaporation over 30 days between the emulsion and hypromellose groups showed significantly less evaporation for the emulsion group (p = 0.001). Lipid layer structure improved from Day 1 to Day 30 of the study with the emulsion group (p = 0.027), but not with the hypromellose group (p = 0.777). The study showed no significant changes in tear production and osmolality with either of the drops. Three subjects in the emulsion group reported adverse effects such as grittiness or blurring lasting 2–5 min, but no changes in the external ocular structure were found (Khanal et al., 2007).

A study compared the effect of four different oils to that of castor oil on WKD‐cells (Said et al., 2007). The cells were incubated with the respective oil for 15 minutes and tested after a 24‐h recovery period. They were tested for viability, proliferation, apoptosis and ROS production. Cells treated with castor oil showed significant necrosis, enhanced intracellular ROS production and activation of the P2X7 cell death receptor. These results show that castor oil might have a cytotoxic effect on conjunctival cells. The four other oils showed no cytotoxic effects (Said et al., 2007).

4.5.2. Sea buckthorn oil

Sea buckthorn (SB) oil is abundant in lipophilic antioxidants and is rich in n‐3 and n‐6 fatty acids (Larmo et al., 2010). A randomized controlled study investigated the effects of an emulsion‐based eyelid spray containing SB oil and SH on dry eye (Larmo et al., 2019). The SB spray resulted in a significant reduction in dry eye symptoms measured by OSDI, questionnaires and daily symptom logs compared with the reference spray or untreated control. Furthermore, no significant adverse effects were observed (Larmo et al., 2019).

4.5.3. Sacha inchi seed oil

Sacha inchi is a plant native to the tropical rain forest of the Amazon region of South America. The chemical constituents found in the plant include lipids (including ω‐3, 6, and 9 fatty acids), proteins, vitamin E (tocopherols), polyphenols, minerals and others (Wang et al., 2018).

A randomized study investigated the safety and efficacy of a multi‐ingredient sacha inchi mircoemulsion (SIME) eye drop compared with a control eye drop containing HA (0.2%) in an isotonic phosphate saline buffer (Laihia et al., 2020). Both treatments were well‐tolerated without adverse events They found that the SIME eye drop significantly improved all three aetiologic factors of dry eye disease (tear film instability, tear hyperosmolarity and cellular damage and inflammation) after 1 month of daily treatment and relieved subjective symptoms, whereas the control HA‐containing eye drop improved only one aetiologic factor significantly (Laihia et al., 2020). The natural antioxidants in sacha inchi seed oil might also provide cytoprotective and wound‐healing activity by attenuating hyperosmolarity‐associated oxidative stress (Laihia et al., 2020).

4.5.4. Other lipids in artificial tears

To our knowledge, studies investigating additional lipids have not been conducted in isolation from other constituents. Future research is needed to demonstrate the safety of these components. However, one study has compared Systane® Balance, which contains mineral oil, dimyristoyl phosphatidylglycerol and sorbitan tristearate, with non‐lipid‐containing Systane® Ultra (Gokul et al., 2018). This randomized, double‐masked trial compared the prophylactic efficacy of Systane® Balance (lipid‐containing) and Systane® Ultra (non‐lipid‐containing) on 30 patients with mild‐to‐moderate dry eye symptoms. One drop of Systane® Balance was applied to one eye and Systane® Ultra was applied simultaneously to the contralateral eye (randomized). Subsequently, the patients were exposed to a standing fan directed towards the eye. Both treatments increased TBUT significantly and prevented a decline below baseline when exposed to the simulated adverse environment (p > 0.05). Only Systane® Balance increased the lipid layer grade significantly and it prevented a decline below baseline post‐adverse environment exposure (p = 0.15). Sixty‐seven per cent of the patients reported greater ocular comfort in the eye receiving Systane® Balance (Gokul et al., 2018).

4.6. Other constituents

4.6.1. Vitamin a palmitate

Vitamin A is the most protective factor of the conjunctival epithelium (Cui et al., 2016). Vitamin A palmitate has been shown to improve goblet cell density. This is shown in a prospective cohort study, in which 23 patients with primary open‐angle glaucoma and seven with normal‐tension glaucoma were enrolled (Cui et al., 2016). The patients were randomized into three groups: vitamin A palmitate eye gel, carbomer eye gel or no additional application. Both treatments significantly improved OSDI scores with no significant inter‐group differences. The authors also showed that the goblet cell density decreased in the non‐intervention group but increased in the vitamin A palmitate group and remained stable in the carbomer group (Cui et al., 2016).

A study by compared the clinical effects of vitamin A palmitate eye gel and SH eye drops in 80 patients with moderate‐to‐severe thyroid‐associated ophthalmopathy with DED (Sun et al., 2023). The patients were randomly assigned into two groups: group A receiving vitamin A palmitate eye gel and group B receiving SH eye drops. Both groups applied the eye drops three times daily. Evaluation of TBUT, Schirmer test, OSFS, OSDI and adverse events took place 1 month after administration. They found that TBUT and OSFS were significantly improved (p < 0.001) in group A. In group B, OSDI and OSFS were significantly improved (p = 0.002). The TBUT value of group A was significantly greater than that of group B (p = 0.009) (Sun et al., 2023).

4.6.2. Disodium edetate

Disodium edetate (traditionally known as EDTA) is a metal chelator, which disrupts the lipopolysac‐charide layer in Gram‐negative bacteria potentiating biocidal, antiseptic or disinfecting effects (Lambert et al., 2004; Maillard & Pascoe, 2024).

A study made by Debbash et al. tested the effect of ethylenediaminetetraacetic acid (EDTA) on WKD‐cells. The concentrations varied between 0.00001% and 0.01%, with the most common concentration used being 0.01% (Debbasch et al., 2001). The study showed that EDTA has much less toxic properties than BAK and CET. In contrast to the quaternary ammonium ions, EDTA did not decrease membrane integrity. Nevertheless, EDTA did show a significant hydrogen peroxide (H₂O₂) production, however, less compared with BAK. Chromatin condensation was observed, but significantly less than observed with the quaternary ammonium compounds (Debbasch et al., 2001).

4.6.3. Phosphate buffer

Artificial tears contain a wide variety of buffers to control pH value including citrate, phosphate and borate buffers (Jones et al., 2017). Approximately one‐third of all medicinal eye drop products in the European Union (EU) use phosphate as part of their buffer system (Agency, 2012). However, rare cases of corneal calcification have been reported in patients using phosphate‐containing eye drops, particularly those with pre‐existing corneal damage. A scientific report has described five cases of corneal calcification following frequent use of a high concentration phosphate buffered artificial tear (Hylo‐Comod with 50.8 mmol/L phosphate). Notably, current formulations of this product no longer contain phosphate (Bernauer et al., 2006).

The Medicines Agency's Committee for Medicinal Products for Human Use (CHMP) reviewed 655 eye drop products, of which 236 contained phosphate and looked at the risk of corneal calcification in patients using phosphate‐containing eye drops (Agency, 2012). Phosphate‐associated calcification was extremely rare (less than 1 reported case per 10.000 bottles distributed) and the evidence does not support a universal recommendation to avoid treatment with eye drops containing phosphate buffer. In addition, calcification is a multifactorial condition, which can occur without using these products. The CHMP considered that the benefits of the phosphate‐containing eye drop medicines authorized in the EU continue to outweigh their risks and the healthcare professionals should consider the individual circumstances of each case (Agency, 2012).

4.6.4. Perfluorohexyloctane

Perfluorohexyloctane (PFHO) is a semifluroinated alkane, which is immiscible with hydrocarbons or water (Vittitow et al., 2023). A prospective, controlled clinical study evaluated the effect of PFHO (Evotears) on higher order aberrations (HOA) after 4 weeks of application four times daily in healthy individuals without a history of DED (Saad & Frings, 2023). The treatment and control group consisted of 104 eyes and 101 eyes, respectively. Measurements were performed with the WASCA aberrometer. They found that administration of a single drop PFHO significantly increased HOA in healthy eyes when applied continuously for 4 weeks (Saad & Frings, 2023). An increase in HOA can potentially lead to undesirable visual disturbances. However, in one of their other studies, they did not demonstrate any significant effect on HOA in the short‐ and long‐term period using a Scheimpflug camera (Pentacam®, Oculus® Optikgeraete GmbH, Wetzlar, Germany) (Habbe et al., 2023). They suspected that the difference in the devices used could explain the variation in the results (Habbe et al., 2023).

Another study showed the potential of PFHO in reducing tear film evaporation (Vittitow et al., 2023). This in vitro study investigated the inhibitory effect of PFHO on tear evaporation, compared with artificial tears and meibum lipids. Key findings reveal that PFHO inhibits evaporation by 82% at 25°C and 88% at 35°C, with a single drop of PFHO reducing evaporation nearly four times more effectively than meibum lipids alone. The study suggests a therapeutic option of PFHO in treating DED (Vittitow et al., 2023).

5. CHOICE OF ARTIFICIAL TEARS

Especially, BAK has been recognized and documented to interfere with the tear film and ocular surface, thereby negatively influencing the progression of DED as well as decreasing compliance. Despite these drawbacks, BAK is still one of the most used preservatives in artificial tears available in the Nordic market (Tables 2 and 3). Cetrimide also exhibited cytotoxic effects in an in vitro study to the same extent as BAK. However, in vivo studies are needed to conclusively determine its cytotoxicity. Research states that alternative preservatives may be cytotoxic as well, although to a lesser extent than BAK, and for some of them, evidence is still lacking. On the other hand, PF artificial tears have proven to be less cytotoxic to the ocular surface, both self‐reported and in vitro. Some drawbacks have been pointed out regarding single‐dose units as they are usually of higher cost, and more difficult to handle especially for patients with tremor or arthritis (Baudouin et al., 2010; Figus et al., 2021; Goldstein et al., 2022; Hedengran et al., 2020). The PF single‐dose containers lack elastic properties causing difficulties with self‐instillation of artificial tears. These factors might lead to decreased adherence, especially for elderly patients and chronic users. The single‐dose package usually contains more than a single dose, which can lead to patients saving them, increasing their risk of infections (Goldstein et al., 2022). Bacteria may alter the pH value of the medication, interfering with the efficacy, why contamination should be avoided (Figus et al., 2021). To avoid this, multi‐dose bottles have been developed with filtration or valve systems hindering penetration of bacteria into the bottle, and liposomal eye sprays have been developed too (Baudouin et al., 2010; Goldstein et al., 2022). Even though it may be more costly to use PF ophthalmic solutions, it is shown that they provide significant benefits, especially for chronic use or vulnerable ocular surfaces (Figus et al., 2021; Kolko et al., 2023; Tong et al., 2012). A 2‐week study compared participant preferences for dispensing systems (unit‐dose vs. multi‐dose) (Pucker et al., 2023). Twenty‐nine participants with digital eye strain were randomized to either PF unit‐dose or multi‐dose Systane Hydration and switched to the counterpart after 1 week. The study showed no clear difference between the dispensing systems as for the majority of the features considered; however, they found that the participants indicated that the multi‐dose system was more environmentally friendly. However, it is important to emphasize that the age group of the participants was/were 28.6 ± 12.0 years, which is younger than the typical DED patient. Furthermore, none of the participants had been diagnosed with a mobility‐related disease. Therefore, the results may not directly apply to older individuals and the healthcare practitioners should discuss options to find the most suitable dispensing system that fits the patient's lifestyle (Pucker et al., 2023).

Although there is a great variety of artificial tears available in the Nordic market, the treatment of dry eyes is not solely a matter of general tear replacement. Different factors should be considered, such as whether the disease is acute or chronic caused by increased evaporation or aqueous deficiency. Artificial tears containing lipids are appropriate for treating evaporative DED, which can be caused by for example meibomian gland dysfunction or mucin deficiency. Aqueous deficient dry eyes can be caused by a reduced lacrimal secretion (Craig et al., 2017; Labetoulle, Benitez‐Del‐Castillo, et al., 2022; Mantelli et al., 2011). In this case, it is important to prescribe artificial tears with viscosity enhancing agents (Labetoulle, Benitez‐Del‐Castillo, et al., 2022). High viscosity artificial tears and gels have a beneficial impact on more severe cases of DED. These artificial tears and gels deliver an increased retention time, which results in transient visual disturbances. This can negatively influence adherence towards treatment and is therefore recommended for overnight use (Pucker et al., 2016). In some cases of DED, the composition of tears might be altered towards a hypertonic tear film. In these cases, hypotonic artificial tears can be beneficial in addressing this aspect of DED (Labetoulle, Benitez‐Del‐Castillo, et al., 2022). Not all cases of DED include hyperosmolarity and inappropriate use of hypotonic artificial tears can potentially disrupt the tear film. The use of hypotonic artificial tears should therefore be directed by an eye care specialist. Besides this, some inactive components should also be considered. Some reports show signs of corneal calcification due to an extensive use of artificial tears with elevated levels of calcium phosphate buffers (Heegaard et al., 2022). Moreover, a significant decrease in cell viability was observed with unpreserved and preserved carbomer 934P in an in vitro study. This cytotoxicity, however, was significantly less than that observed with BAK alone. Castor oil also showed in vitro cytotoxic effects; however, no in vivo studies yielded similar results. Besides this, no cytotoxic effects are, to our knowledge, known to the other constituents.

Several types of eye drops may be prescribed, but the majority are available for purchase over the counter (Pucker et al., 2016). Individuals dealing with chronic conditions, often requiring a substantial quantity of medication, commonly seek financial assistance for their prescriptions. This support is only applicable for artificial tears that need a prescription. Out of the 11 eye drops that contain BAK or CET, six are available by prescription, making them an option for chronic patients, which leads to a problematic issue. Thus, it is crucial for a physician to prescribe suitable artificial tears or provide instructions to the patients regarding the most appropriate artificial tears. Importantly, patients with chronic dry eyes who are long‐term users of artificial tears are the most vulnerable patient group due to increased sensitivity of the ocular surface. In this sense, eliminating adverse effects is crucial and may lead to enhanced adherence to therapy.

6. CONCLUSION

A total of 88 artificial tears were identified in the Nordic market. Among these, 28 contained one or more preservatives, representing a substantial portion of the available alternatives. Eight different preservatives were found in the Nordic market: BAK, CET, PQ‐1, Purite®, OcuPure®, Oxyd®, PHMB, and phenoxyethanol. BAK has been associated with consistent ocular toxicity in both preclinical and clinical studies. This contributes to the vicious cycle of adverse effects in patients and may aggravate DED as well as decrease adherence. Alternative preservatives such as PQ‐1 and Purite® have been shown to be less toxic compared with BAK, yet evidence suggests that these alternatives might be detrimental as well, especially for CET. Moreover, adequate scientific evaluation regarding preservative agents such as PHMB and OcuPure® is lacking. Preservative‐free artificial tears have demonstrated an enhanced safety profile to the ocular surface, both self‐reported and in vitro. Studies show that especially chronic patients, who are long‐term users of artificial tears, should avoid preserved artificial tears, especially when preserved with BAK and CET. Studies also showed increased patient satisfaction and greater resistance towards environmental stress, when receiving lipid‐containing artificial tears. To avoid adverse effects, the constituents and the toxicity profile of artificial tears must be considered by healthcare professionals when prescribing or recommending artificial tears to their patients.

Mikha, K.N. , Rasmussen, C.E.Ø. , Ahrensberg, S.N.G. , Freiberg, J. , Wozniak, R.A. , Alnoor, U.M.K. et al. (2025) Can the choice of artificial tears harm patients? A narrative review with an overview of the Nordic market. Acta Ophthalmologica, 103, 586–608. Available from: 10.1111/aos.17455