Introduction

Patients who are admitted to Intensive Care Units (ICUs) usually experience failures in one or more vital organ systems such as respiratory, neurologic, and cardiovascular systems accompanied by many other complications. Ocular Surface Disease (OSD) is a common complication in this patient population, but receive minimal clinical consideration to prevent it [1, 2]. Due to the ICU patients’ life-threatening conditions, the eye care and symptoms of the OSD are often given low priority. OSD indicates damage to the surface layers of the eye, specifically, the corneal epithelial and conjunctiva, which is clinically diagnosed based on the slit lamp examination using different dyes such as fluorescein [3]. OSD is associated with different risk factors, with the most common being impaired secretion and distribution of tear film over the ocular surface, altered tear components concentration due to lack of blinking, incomplete lid closure (lagophthalmos), and lid or conjunctival edema occurring with the use of ventilator [4–9]. There is a range of eye complications from mild Conjunctival infections to severe ocular damage. Keratopathy (3.6% to 60%), chemosis (9% to 80%), and microbial keratitis are the most prevalent ocular disorders in ICU patients [10]. Insufficient eye care of ICU patients can increase the risk of serious eye problems such as corneal abrasions and ulceration, infectious keratitis, and even corneal perforation and loss of vision [11, 12].

It is highly recommended to integrate eye care into the treatment plan for critically ill patients in order to protect their eyes from OSDs. In recent decades, many evidence-based practices for eye care were developed; these practices include a wide range of interventions such as 1) washing the eyes with normal saline solution; 2) using lubricating ointments or teardrops; 3) moisturizing the chambers by applying a polyethylene eye cover, swimming goggles; 4) covering the eyes with eye shields, pads, or patches; and 5) closing the eyelid with closure by transparent tape or tarsorrhaphy. There is no clear evidence as to which of these interventions represent the ideal eye care for ICU patients to prevent OSD [4, 5, 9, 11, 13–18]. Accordingly, healthcare professionals’ beliefs and traditions have played a major role in selecting eye care interventions in the ICU settings in different countries [5, 17]. Applying polyethylene eye covers has been suggested as the most effective intervention [5] and the main components of the evidence-based protocol to prevent exposure keratopathy in the ICU settings [19]. But, in a very recent study of eye care for the intensive care unit, the lubricants and taping of the lids are only advised for the conjunctival and corneal exposure [18]. The polyethylene eye cover was not considered as a part of the standard eye care for critically ill patients as well. Inconsiderately, only has been reported a daily washing of eyelids with normal saline and sterile gauze and application of ocular lubricants at least twice per day as the standard care in ICUs [18].

In Iran, washing out eye discharge with sterile normal saline is considered as part of routine eye care for ICU patients, but this intervention is not well supported empirically and is no longer recommended worldwide. Our research team collaborated with health professionals to select alternative eye care interventions to prevent OSD in ICU patients. The team members carefully reviewed empirical and clinical literature to determine the effectiveness or benefits and risks of alternative interventions, while also considering their feasibility in the local ICU context. The results of this critical appraisal indicated the following: 1) Irrigating eye with sterile normal saline could cause an increase in the incidence of OSD and cross-infection, and the cotton wool balls can cause corneal scratching [5]. 2) lubrication materials could be hard and time-consuming to apply on regular or frequent time intervals by nurses [20]; further, the ointment containers can be infected if not used properly. 3) Using eyelid sutures makes regular eye checkups difficult; the sutures are also unpleasant in appearance and block the eye activities [20]. 4) Tarsorrhaphy is also invasive and almost impossible to apply to our lagophthalmos patients. 5) Moisture chamber goggles are only designed for adults and are inconvenient to use when patients are in a supine position [20]. Securing the moisture chamber goggles around the head also increases the risk of edematous eyes in ventilated patients. 6) Mechanical closure of the eyes is recommended in the case of lagophthalmos [21]. 7) Polyethylene covers may be pulled by infants, children, or agitated patients [20]. Lastly, eye covering is claimed as the most effective intervention in preventing exposure keratopathy [5].

After examining the benefits and risks of each eye care interventions and its feasibility in our ICU settings, the research team decided to use polyethylene eye covers. However, evidence to support their clinical efficacy in reducing the incidence and severity of OSD is limited.

Materials and methods

Study design

This was a prospective multicenter randomized triple-blinded three-arm clinical trial. After considering three groups where every patient may be allocated, to control for possible between-participant differences in clinical conditions that may influence the eyes’ status, randomization was done at two levels. First, eye care interventions were randomly allocated to one of the left or right eye for each participant in each group, by tossing a coin as detailed next.

• Group “A” patients received artificial teardrops for the left eyes and normal saline for the right eyes.
• Group “B” patients received polyethylene cover for the left eyes and normal saline for the right eyes.
• Group “C” patients received polyethylene cover to the left eyes and artificial teardrops to the right eyes.

Second, eligible participants were assigned to one of three treatment groups by permuted block randomization (AABBCC, ABCABC, BBCCAA …) enveloped in the opaque packets and selected by a nurse outside the study. The sample consisted of patients newly admitted to the participating ICUs. Patients were enlisted if only they were over 18 years of age, comatose (with a Glasgow Coma Scale (GCS) less than or equal to 8 of 15), mechanical ventilated and who had no trauma to the areas of the face and eyes and no history of chronic disease, electrolyte imbalance, cataract surgery or glaucoma. An ophthalmologist confirmed that the patients had a lack/ difficulty in blinking (less than five times per minute), incomplete eyelid closure, and a healthy cornea after examination by the fluorescein and a portable slit lamp. Patients who showed signs of consciousness or natural blinking reflexes required cardiopulmonary resuscitation during the trial and got discharged from ICUs to the other centers or died, were excluded.

As the patients were comatose, their blinding did not apply. A single observer who was blinded to the patient assignments evaluated the patients’ eyes based on the Corneal Fluorescein Staining Pattern. The data analyst blinded to the allocated treatment analyzed the primary outcome.

The sample size was calculated according to the formula below. With an 80% power and a 2-sided significance level of α = .05, the number of people needed in each group was 25. Anticipating a total attrition rate of 20%, a total of 90 patients were selected and included in the study, 30 patients for each arm.

Of the 90 eligible patients, 79 patients completed the trial participation. The number of patients in group A was 25, group B came to 29, and group C remained 25. In a post hoc power analysis by G*Power 3.1.9.2 software, when we set a small effect size of 0.25, a significance level of α = .05, with a total sample size of 158 in three groups, the achieved power computed more than 0.80.

Results

In this study, the data of 158 eyes were compared, 54 eyes for normal saline, 50 for artificial teardrops, and 54 for polyethylene cover interventions (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. CONSORT flow diagram.

https://doi.org/10.1371/journal.pone.0248830.g001

Patients assigned to the three treatment groups were comparable on demographic and clinical characteristics (all p’s > .05) (Table 1). The majority of patients were older men, admitted for the management of Neurology diseases.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Comparison of the studied demographic characteristics among groups (n = 90).

https://doi.org/10.1371/journal.pone.0248830.t001

Incidence of ocular surface disease

The presence of OSD in each eye, in patients assigned to the three treatment groups, is shown in Table 2. The OSDs were observed in 65 (41.14%) out of 158 eyes examined. McNemar’s test compared the presence of OSD across groups. The results indicated that in group A, the artificial teardrops intervention was considerably more useful than the sterile normal saline drops intervention (p< .01). In group B, the polyethylene cover intervention was significantly better than the sterile normal saline drops intervention (p< .001). Finally, in group C, the polyethylene cover intervention was also better than the artificial teardrops intervention (p< .01).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Comparison of the patients’ eyes according to the incidence of the ocular surface disease.

https://doi.org/10.1371/journal.pone.0248830.t002

Severity of ocular surface disease

The severity grading of the OSDs according to the Corneal Fluorescein Staining Pattern in three treatment groups is shown in Table 3. Although the artificial teardrops intervention was better than normal saline drops intervention, Kruskal–Wallis test showed that the polyethylene eye covers intervention was the best in reducing the severity of OSD (p < .001).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Comparison of the patients’ eyes according to the severity of the ocular surface disease.

https://doi.org/10.1371/journal.pone.0248830.t003

Discussion

In this clinical trial, three eye care interventions, applied over a five-day period were compared. The findings showed that the artificial teardrops intervention is more effective than the normal saline drops intervention, and the polyethylene eye cover is the best intervention for reducing the incidence and severity of OSD in comatose patients. Moreover, eye covering with a transparent thin plastic film was not a barrier for frequent observation and monitoring of the cornea. Our results reconfirmed the effectiveness of the polyethylene eye-cover. Granting that polyethylene covering is recommended as an effective eye-care in ICU patients [28], the method of using it is not explained in detail. This is the first study for the ICU patient’s eye care in which the eye covering was precisely defined, every 12 hours with considering a 15-minute of the cornea oxygenation in each eye re-covering. So, our procedure of polyethylene eye covering is able to improve the evidence-based eye-care in ICU patients.

Our findings are also consistent with those reported by Shan and Min in showing that polyethylene eye covers maintain a healthy cornea; the polyethylene cover fared better than the moist chamber or artificial teardrops [29]. Using polyethylene eye covers every 12 hours has also been compared to Carbomer eye drops applied every six hours for a period of five days; the polyethylene cover was found to significantly prevent the dry eye syndrome in unconscious patients [30]. Also, polyethylene covers were significantly more beneficial in the prevention of keratopathy than Liposic ophthalmic ointment and artificial teardrops [31]. Polyethylene eye covers are more effective than the hypromellose and Lacri-Lube combination in reducing the incidence of corneal ulceration in intensive care patients [32].

The effectiveness of polyethylene covers is not only related to the simple eye closing mechanism. The results of a study showed that using wet gauze to cover the eye was not significantly useful in preventing dry eye and corneal ulcers in patients during general anesthesia [7]. Taping eyes is another method of eye closing that was reported to be more effective than the eye gel membrane in preventing the development of keratopathy in comatose patients. Taping the eye and eyelid has been advocated as a safe prophylactic method, even for patients who require frequent pupillary examination [33]; however, the risk of injury to the eyelids and nearby skin and the probability of increasing the relatives’ distress remain [34].

Our study results support the usefulness of covering eyes with a polyethylene film in the prevention of OSD; polyethylene eye covers simultaneously protect eyes by creating a moist chamber, provide a naturally lubricated eye surface by preventing the evaporation of tears, and stop the growth of microorganisms in patient’s eyes, especially during airway suctioning and tube feeding procedures. In a comparison of the two eye care interventions, polyacrylamide hydrogel dressings (Geliperm) versus an ocular lubricant (Lacrilube), Geliperm was found as effective as Lacrilube in preventing exposure keratopathy in ICU patients [35]. The polyacrylamide hydrogel dressings provide a moist environment at the surface of the wound, are resistant to bacteria, but permeable to water vapor, gases, and small protein molecules. Although these dressings are a means for eye closing, their permeability makes them an ocular lubricant in the eye care of comatose patients.

Using polyethylene eye cover can minimize the frequency of eye cleansing by swab wash. Eye irrigation with sterile water/normal saline-soaked cotton balls or gauze is typically recommended as a hygiene regimen or regular cleansing of the eye for critical patients in ICUs [18, 36], but eye irrigation increases the incidence of OSD and infection, and corneal scratching [5]. Covering eyes with polyethylene every 12 hours can save time on care and decrease threats of eye abrasion and infection. In our study, the eyes covered with polyethylene have usually remained clean after 12 hours of the previous cover exchange.

The finding of this clinical trial and of another study indicate that polyethylene covers are effective, time-saving, and easy intervention for eye care in critically ill patients [29]. So, it is seriously recommended to the ICU nurses to substitute applying normal saline drops/irrigation with polyethylene eye covering. The limitation of this study relates to the evaluation of symptoms of OSD in comatose patients. Further research is needed to examine the eye care intervention in conscious patients who are not able to close their eyelids well, such as Bell’s palsy, severe exophthalmos, and severe skin disorders as ichthyosis. Research with this patient population would provide a good opportunity to inquire about patients’ perspectives on different eye care interventions. In this study, we measured the incidence and severity of the punctate epithelial erosions of the cornea. So, the other forms of OSDs in ICU patients such as epithelial defects, infections, impaired secretion and distribution of tear film over the ocular surface, tear components concentration, and conjunctival manifestations due to the use of the polyethylene eye covering and/or artificial teardrops need more studies. Although our eligible patients had incomplete eyelids closure and each patient was a member of the control and/or intervention group simultaneously, the grades of eye closure were not our main criterion. So, more studies are needed about the only cornea exposed eyes for more specific results. Future studies could also evaluate the initiative to implement pre-disposable polyethylene eye covers into nursing practice. A cost-effectiveness analysis may also be of relevance, given the decreased financial costs and less timewasting with the eye covering.

Conclusion

In conclusion, Polyethylene eye cover demonstrated superior efficacy in preventing both incidence and severity of the OSD in comatose patients compared to normal saline and artificial teardrops. Adding a polyethylene eye cover to the routine care in critical care settings is necessary to prevent the OSD and subsequent complications.

Supporting information

Acknowledgments

This study is derived from a master thesis in critical care nursing. The authors kindly acknowledged the Research and Technology Vice-chancellor at Hamadan University of Medical Sciences. The authors are extremely grateful and sincerely acknowledge all the hospital personnel, and patients and their families contributed to this research.