- Email: [email protected]
Corneal Injury from Presurgical Chlorhexidine Skin Preparation
Accepted Manuscript Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation Gregory Bever, MD, Frank Brodie, MD, David G. Hwang, MD, FACS PII:
S1878-8750(16)30861-0
DOI:
10.1016/j.wneu.2016.09.035
Reference:
WNEU 4575
To appear in:
World Neurosurgery
Received Date: 24 July 2016 Revised Date:
7 September 2016
Accepted Date: 8 September 2016
Please cite this article as: Bever G, Brodie F, Hwang DG, Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation, World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.09.035. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation Gregory Bever MD1, Frank Brodie MD1, David G. Hwang MD, FACS1
1 Department
SC
RI PT
of Ophthalmology and Cornea Service, University of California San Francisco, San Francisco, CA
M AN U
Support from That Man May See and Research to Prevent Blindness
Running headline: Corneal Injury from Pre-surgical Chlorhexidine
Word count: 1516
TE D
Skin Preparation
To whom correspondence should be addressed: David G. Hwang MD, FACS; Cornea Service; 10 Koret Way; San Francisco, CA 94143-0730
AC C
EP
Email: [email protected]
ACCEPTED MANUSCRIPT
Abstract
AC C
EP
TE D
M AN U
SC
RI PT
Chlorhexidine skin preparation has been shown to provide highly effective antimicrobial pre-surgical skin cleansing. However, there is a significant risk of ocular toxicity when it is used in peri-ocular areas. We describe two cases of significant corneal damage from 4% chlorhexidine gluconate pre-operative skin cleanser, despite use of protective occlusive dressing over the eyes. Because of chlorhexidine's potential for severe corneal toxicity, alternative agents such as 10% povidone-iodine should be considered for skin preparation near the periocular areas whenever possible. If chlorhexidine gluconate must be employed in such settings, great care must be exercised to avoid contact with the eyes and additional protective measures (e.g., absorbent eye pads in association with tightly occlusive dressings) used whenever possible.
ACCEPTED MANUSCRIPT
Introduction
M AN U
SC
RI PT
The benefits of preoperative use of chlorhexidine skin scrub for decreasing the incidence of surgical site infection have been shown in a large randomized controlled trial1 and echoed in a review and meta-analysis of the literature.2 However, if chlorhexidine inadvertently comes into contact with the ocular surface, progressive corneal damage can occur in a dose-dependent fashion.3,4 At lower concentrations, chlorhexidine can cause toxic effects on the epithelial layers of the cornea and conjunctiva, which can cause transient epithelial disruption and abrasion. Typically, full recovery without lasting visual sequelae is possible.3 However, at higher concentrations, irreversible injury occurs to the corneal endothelial monolayer, which in humans lacks the ability to regenerate.3,4 With sufficient loss of endothelial cells, corneal edema ensues and the cornea loses its transparency; the resultant visual loss can be severe. While lesser degrees of corneal endothelial injury may heal over a period of weeks to months, severe corneal endothelial injury will result in irreversible corneal edema and accompanying visual loss that can only be treated by corneal transplantation.
EP
TE D
There have been several reports of ocular toxicity occurring as a result of inadvertent exposure of the cornea to chlorhexidine applied to the skin as preparation for nonocular surgery. Phinney et al5 describe a series of patients undergoing nonocular surgery who had ocular exposure to Hibiclens (4% chlorhexidine gluconate, 4% iso-propyl alcohol, purified water) during preoperative skin cleaning. All patients suffered corneal abrasion and corneal edema with reduction in visual acuity. In two patients the corneal edema did not respond to treatment with corticosteroid eye drops and ultimately required corneal transplantation. Hamed et al6 describe a case of a pediatric patient who underwent Hibiclens skin preparation immediately prior to a craniotomy procedure; by the end of the case, the cornea was found to be completely opacified on the operative side with a complete corneal abrasion. Recently Liu et al7 describe a case of a patient who underwent resection of gingival cancer with inadvertent ocular exposure to chlorhexidine skin preparation, leading to persistent corneal edema that eventually required corneal transplantation.
AC C
We report two neurosurgical patients who experienced severe corneal toxicity and visual loss due to presumed inadvertent topical ocular exposure to the chlorhexidine skin preparation.
Case #1
A 61-year-old man with previously normal vision presented for ophthalmological evaluation one day after undergoing an uncomplicated posterior cervical laminectomy, decompression, and arthrodesis with autograft from C3-C7. The patient had been in the prone position throughout the procedure and underwent a ten-minute surgical prep using Exidine 4% (chlorhexidine gluconate,
ACCEPTED MANUSCRIPT
CareFusion, Vernon Hills, IL) with protective patching of the eyes using a transparent nylon breathable membrane (Tegaderm™, 3M Corporation, Maplewood, MN). Seven hours elapsed between application of the surgical prep solution and the end of the surgery.
SC
RI PT
Upon presentation to the ophthalmology clinic one day later, the patient was found to have a 95% epithelial defect of his left eye without evidence of infection. The epithelial defect was persistent over the next two and a half months despite varied and extensive treatment including bandage contact lenses, topical corticosteroid drops, lubricating drops, topical antimicrobial prophylaxis, and suture closure of the eyelids (tarsorrhaphy). The patient was in persistent pain throughout the course of these unsuccessful treatments. Although the cornea never appeared infected, cultures were taken and bacterial, fungal, and herpesvirus infections were ruled out.
EP
Case #2
TE D
M AN U
Although the corneal epithelial defect healed after two and half months, the cornea still exhibited evidence of considerable inflammation including significant stromal edema with Descemet’s membrane folds, stromal hemorrhage, and deep stromal neovascularization approaching the visual axis. By four months after the initial neurosurgical procedure, the patient had significant stromal scarring with neovascularization, increased corneal thickness, and a visual acuity of 20/150. These findings persisted and at postoperative month six, the patient was referred for consideration of corneal transplantation (Fig. 1). Specular microscopy confirmed endothelial toxicity (Fig. 2). The clinical presentation and onset immediately after the neurosurgical procedure strongly suggested that the patient had suffered severe ocular surface toxicity from inadvertent chlorhexidine exposure, despite the application over the closed eyes of a transparent, semipermeable nylon adhesive membranes prior to performing chlorhexidine preparation of the head and neck.
AC C
A 46-year-old woman underwent an uncomplicated left pterional craniotomy with orbital osteotomy and anterior clinoidectomy for resection of a left clinoidal meningioma. She was placed in the supine position and after occlusion of the eyes with a transparent, adhesive nylon membrane (Tegaderm™, 3M Corporation, Maplewood, MN) received a ten-minute surgical skin preparation of Exidine 4% (chlorhexidine gluconate, CareFusion, Vernon Hills, IL) to the head. Ten hours elapsed between application of the surgical prep solution and the end of the surgery. After waking from successful resection of the meningioma, the patient reported eye pain ipsilateral to the operative left side. Upon bedside ophthalmic examination, the patient’s left eye was noted to have a large inferior hemicorneal epithelial defect with a well demarcated horizontal superior border corresponding to the upper eyelid margin. This was initially presumed to be a post-traumatic or exposure-related injury, and the patient was treated with erythromycin ophthalmic
ACCEPTED MANUSCRIPT
Discussion
TE D
M AN U
SC
RI PT
ointment applied to the affected eye. By postoperative day 13, there was complete resolution of the corneal abrasion but two months later the patient returned with complaints of ocular pain and decreased acuity on the operative side. Best corrected visual acuity was 20/50 in the affected left eye and 20/20 without correction in the fellow right eye. Intraocular pressure was normal in both the affected and fellow eyes, at 14 mm Hg and 13 mm Hg respectively (normal range, 10-21 mm Hg). Slit lamp examination showed sectoral microcystic corneal edema in the same distribution as the previous postoperative epithelial defect. The differential diagnosis included herpetic keratitis, so the patient was empirically treated with oral valacyclovir and a topical corticosteroid. However, after several weeks of treatment and lack of clinical response, the patient was sent to a corneal surgeon for evaluation. The corneal lesion’s interpalpebral location, sharply demarcated horizontal superior border, and depth to the corneal endothelium (Fig. 3) were all consistent with a toxic, penetrant substance gaining entry to the ocular surface via a partially open eyelid fissure but sparing the superior cornea where the eyelid was in apposition to the cornea. Endothelial toxicity was confirmed by direct imaging using specular microscopy, which demonstrated marked loss of endothelial cells (Fig. 4). Toxicity from the chlorhexidine-based surgical preparation was identified as the likely culprit given the onset of injury immediately after the mengioma resection. At present, the patient is being managed with conservative topical therapy but given extensive endothelial damage, the patient will likely need corneal transplantation in the future.
AC C
EP
Ocular injuries are infrequent following nonocular surgical procedures; one large retrospective study showed an incidence of about 0.056%.8 Anesthesiologists take various precautions including eyelid taping and bio-occlusive dressing (such as Tegaderm™) application to prevent ocular damage.9 Unlike the cases in this report, however, such ocular injuries are generally superficial, resulting in corneal epithelial disruption, and usually heal completely within 24-48 hours without specific therapy. The corneal epithelial defects in these cases took two and half months and 13 days, respectively, to heal and even after the corneal epithelial layer healed there was persistent damage to the deeper corneal layers, including the stroma and endothelium. In each of the above cases, the ocular damage caused by chlorhexidine was severe, causing protracted pain, visual loss lasting months, and a strong likelihood that corneal transplantation to restore good vision would be required. Recent data in the surgical literature has led to a shift toward using chlorhexidine rather than povidone-iodine in surgical prep because of reduced rates of surgical site infections associated with the former agent.1 While povidone-iodine at low concentrations is routinely and safely applied directly to the ocular surface for surgical prophylaxis in ophthalmic surgery, chlorhexidine has long been shown to cause considerable damage to ocular structures. 3 The injuries described range
ACCEPTED MANUSCRIPT
from epithelial defects that self resolve without sequelae, to irreversible opacification of the cornea requiring full-thickness corneal transplantation.
M AN U
SC
RI PT
The purpose of this report is to bring awareness to the potential ocular risks associated with chlorhexidine surgical prep, particularly for neurosurgical procedures. Surgical procedures calling for a prone position and those requiring preparation near the upper neck and face are at especially high risk for ocular injury resulting from inadvertent seepage of the chlorhexidine into the eyes.10 In both cases reported here, attempts were made to protect the eyes with a transparent adhesive nylon membrane (Tegaderm™), yet injury still occurred. We speculate that the porous, breathable nature of the transparent nylon dressing may have allowed the cationic chlorhexidine to freely permeate through the membrane into the underlying eye. It is also possible the occlusive nylon adhesive membrane did not remain fully adherent throughout the length of the procedure. Additionally, the duration of surgery may have played a role, since longer surgical cases have been associated with increased risk for ocular injury from inadvertent chlorhexidine exposure.8 The two patients in this report had surgical case lengths of seven hours and ten hours, respectively; that extended period of possible chlorhexidine exposure to the ocular surface could have contributed to the severe damage observed.
AC C
EP
TE D
The significant long-term patient consequences described in this report are reminders that special attention should be paid to protecting the eyes from exposure when employing chlorhexidine skin preparation in high-risk cases. Additionally, pre-operative discussion of surgical risks should include possible toxicity from surgical scrubs balanced against risk of surgical site infections. Measures that may lessen such risks include consideration of using povidone-iodine skin scrubs as an alternative to chlorhexidine in high-risk cases, avoiding applying the chlorhexidine preparation to the periocular skin, and placement of an absorbent eye pad over the closed eyes before applying the occlusive dressing, to serve as an additional layer of protection against seepage of chlorhexidine through or around the dressing.
ACCEPTED MANUSCRIPT
Darouiche RO, Wall MJJ, Itani KMF, et al. Chlorhexidine–Alcohol versus Povidone–Iodine for Surgical-Site Antisepsis. N Engl J Med. 2010;362(1):18-26. doi:10.1056/NEJMoa0810988.
2.
Lee I, Agarwal RK, Lee BY, Fishman NO, Umscheid CA. Systematic review and cost analysis comparing use of chlorhexidine with use of iodine for preoperative skin antisepsis to prevent surgical site infection. Infect Control Hosp Epidemiol. 2010;31(12):1219-1229. doi:10.1086/657134.
3.
Green K, Livingston V, Bowman K, Hull DS. CHlorhexidine effects on corneal epithelium and endothelium. Arch Ophthalmol. 1980;98(7):1273-1278. doi:10.1001/archopht.1980.01020040125020.
4.
Mac Rae SM, Brown B, Edelhauser HF. The corneal toxicity of presurgical skin antiseptics. Am J Ophthalmol. 1984;97(2):221-232.
5.
Phinney RB, Mondino BJ, Hofbauer JD, et al. Corneal Edema Related to Accidental Hibiclens Exposure. Am J Ophthalmol. 1988;106(2):210-215. doi:10.1016/0002-9394(88)90837-9.
6.
Hamed LM, Ellis FD, Boudreault G, Ii FMW, Helveston EM. Hibiclens Keratitis. Am J Ophthalmol. 1987;104(1):50-56. doi:10.1016/0002-9394(87)90292-3.
7.
Liu H-Y, Yeh P-T, Kuo K-T, Huang J-Y, Lin C-P, Hou Y-C. Toxic Keratopathy Following the Use of Alcohol-Containing Antiseptics in Nonocular Surgery. JAMA Ophthalmol. 2016;134(4):449. doi:10.1001/jamaophthalmol.2016.0001.
8.
Roth S, Thisted RA, Erickson JP, Black S, Schreider BD. Eye injuries after nonocular surgery. A study of 60,965 anesthetics from 1988 to 1992. Anesthesiology. 1996;85(5):1020-1027.
9.
Grixti A, Sadri M, Watts MT. Corneal Protection during General Anesthesia for Nonocular Surgery. Ocul Surf. 2013;11(2):109-118. doi:10.1016/j.jtos.2012.10.003.
EP
TE D
M AN U
SC
RI PT
1.
AC C
10. Yu H-D, Chou A-H, Yang M-W, Chang C-J. An Analysis of Perioperative Eye Injuries After Nonocular Surgery. Acta Anaesthesiol Taiwan. 2010;48(3):122129. doi:10.1016/S1875-4597(10)60043-4.
ACCEPTED MANUSCRIPT
SC
RI PT
Fig. 1. Case 1. Slit lamp photograph of the left cornea eight months after neurosurgical procedure showing significant opacification involving the visual axis.
M AN U
Fig. 2. Case 1. A: Specular microscopy of unaffected right eye showing normal endothelial cell density, 2427 cells/mm2. B: Specular microscopy of damaged left eye showing abnormal morphology and quantity (910 cells/mm2) of endothelial cells. These damaged endothelial cells do not regenerate.
EP
TE D
Fig. 3. Case 2. Slit lamp photograph of the affected left eye showing corneal haze.
AC C
Fig. 4. Case 2. A: Specular microscopy of unaffected right eye showing normal endothelial cell density, 2941 cells/mm2. B: Specular microscopy of damaged left eye showing abnormal morphology and quantity (518 cells/mm2) of endothelial cells. These damaged endothelial cells do not regenerate.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation Gregory Bever MD, Frank Brodie MD, David G. Hwang MD, FACS
•
Chlorhexidine 2% pre-surgical skin preparation used on the face in two neurosurgical cases led to permanent corneal damage Alternatives to Chlorhexidine include povidone-iodine 10% which does not have permanent corneal toxicity
AC C
EP
TE D
M AN U
SC
•
RI PT
Highlights
ACCEPTED MANUSCRIPT
Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation Gregory Bever MD, Frank Brodie MD, David G. Hwang MD, FACS
RI PT
Abbreviations
AC C
EP
TE D
M AN U
SC
1) mm Hg: millimeters of mercury
S1878-8750(16)30861-0
DOI:
10.1016/j.wneu.2016.09.035
Reference:
WNEU 4575
To appear in:
World Neurosurgery
Received Date: 24 July 2016 Revised Date:
7 September 2016
Accepted Date: 8 September 2016
Please cite this article as: Bever G, Brodie F, Hwang DG, Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation, World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.09.035. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation Gregory Bever MD1, Frank Brodie MD1, David G. Hwang MD, FACS1
1 Department
SC
RI PT
of Ophthalmology and Cornea Service, University of California San Francisco, San Francisco, CA
M AN U
Support from That Man May See and Research to Prevent Blindness
Running headline: Corneal Injury from Pre-surgical Chlorhexidine
Word count: 1516
TE D
Skin Preparation
To whom correspondence should be addressed: David G. Hwang MD, FACS; Cornea Service; 10 Koret Way; San Francisco, CA 94143-0730
AC C
EP
Email: [email protected]
ACCEPTED MANUSCRIPT
Abstract
AC C
EP
TE D
M AN U
SC
RI PT
Chlorhexidine skin preparation has been shown to provide highly effective antimicrobial pre-surgical skin cleansing. However, there is a significant risk of ocular toxicity when it is used in peri-ocular areas. We describe two cases of significant corneal damage from 4% chlorhexidine gluconate pre-operative skin cleanser, despite use of protective occlusive dressing over the eyes. Because of chlorhexidine's potential for severe corneal toxicity, alternative agents such as 10% povidone-iodine should be considered for skin preparation near the periocular areas whenever possible. If chlorhexidine gluconate must be employed in such settings, great care must be exercised to avoid contact with the eyes and additional protective measures (e.g., absorbent eye pads in association with tightly occlusive dressings) used whenever possible.
ACCEPTED MANUSCRIPT
Introduction
M AN U
SC
RI PT
The benefits of preoperative use of chlorhexidine skin scrub for decreasing the incidence of surgical site infection have been shown in a large randomized controlled trial1 and echoed in a review and meta-analysis of the literature.2 However, if chlorhexidine inadvertently comes into contact with the ocular surface, progressive corneal damage can occur in a dose-dependent fashion.3,4 At lower concentrations, chlorhexidine can cause toxic effects on the epithelial layers of the cornea and conjunctiva, which can cause transient epithelial disruption and abrasion. Typically, full recovery without lasting visual sequelae is possible.3 However, at higher concentrations, irreversible injury occurs to the corneal endothelial monolayer, which in humans lacks the ability to regenerate.3,4 With sufficient loss of endothelial cells, corneal edema ensues and the cornea loses its transparency; the resultant visual loss can be severe. While lesser degrees of corneal endothelial injury may heal over a period of weeks to months, severe corneal endothelial injury will result in irreversible corneal edema and accompanying visual loss that can only be treated by corneal transplantation.
EP
TE D
There have been several reports of ocular toxicity occurring as a result of inadvertent exposure of the cornea to chlorhexidine applied to the skin as preparation for nonocular surgery. Phinney et al5 describe a series of patients undergoing nonocular surgery who had ocular exposure to Hibiclens (4% chlorhexidine gluconate, 4% iso-propyl alcohol, purified water) during preoperative skin cleaning. All patients suffered corneal abrasion and corneal edema with reduction in visual acuity. In two patients the corneal edema did not respond to treatment with corticosteroid eye drops and ultimately required corneal transplantation. Hamed et al6 describe a case of a pediatric patient who underwent Hibiclens skin preparation immediately prior to a craniotomy procedure; by the end of the case, the cornea was found to be completely opacified on the operative side with a complete corneal abrasion. Recently Liu et al7 describe a case of a patient who underwent resection of gingival cancer with inadvertent ocular exposure to chlorhexidine skin preparation, leading to persistent corneal edema that eventually required corneal transplantation.
AC C
We report two neurosurgical patients who experienced severe corneal toxicity and visual loss due to presumed inadvertent topical ocular exposure to the chlorhexidine skin preparation.
Case #1
A 61-year-old man with previously normal vision presented for ophthalmological evaluation one day after undergoing an uncomplicated posterior cervical laminectomy, decompression, and arthrodesis with autograft from C3-C7. The patient had been in the prone position throughout the procedure and underwent a ten-minute surgical prep using Exidine 4% (chlorhexidine gluconate,
ACCEPTED MANUSCRIPT
CareFusion, Vernon Hills, IL) with protective patching of the eyes using a transparent nylon breathable membrane (Tegaderm™, 3M Corporation, Maplewood, MN). Seven hours elapsed between application of the surgical prep solution and the end of the surgery.
SC
RI PT
Upon presentation to the ophthalmology clinic one day later, the patient was found to have a 95% epithelial defect of his left eye without evidence of infection. The epithelial defect was persistent over the next two and a half months despite varied and extensive treatment including bandage contact lenses, topical corticosteroid drops, lubricating drops, topical antimicrobial prophylaxis, and suture closure of the eyelids (tarsorrhaphy). The patient was in persistent pain throughout the course of these unsuccessful treatments. Although the cornea never appeared infected, cultures were taken and bacterial, fungal, and herpesvirus infections were ruled out.
EP
Case #2
TE D
M AN U
Although the corneal epithelial defect healed after two and half months, the cornea still exhibited evidence of considerable inflammation including significant stromal edema with Descemet’s membrane folds, stromal hemorrhage, and deep stromal neovascularization approaching the visual axis. By four months after the initial neurosurgical procedure, the patient had significant stromal scarring with neovascularization, increased corneal thickness, and a visual acuity of 20/150. These findings persisted and at postoperative month six, the patient was referred for consideration of corneal transplantation (Fig. 1). Specular microscopy confirmed endothelial toxicity (Fig. 2). The clinical presentation and onset immediately after the neurosurgical procedure strongly suggested that the patient had suffered severe ocular surface toxicity from inadvertent chlorhexidine exposure, despite the application over the closed eyes of a transparent, semipermeable nylon adhesive membranes prior to performing chlorhexidine preparation of the head and neck.
AC C
A 46-year-old woman underwent an uncomplicated left pterional craniotomy with orbital osteotomy and anterior clinoidectomy for resection of a left clinoidal meningioma. She was placed in the supine position and after occlusion of the eyes with a transparent, adhesive nylon membrane (Tegaderm™, 3M Corporation, Maplewood, MN) received a ten-minute surgical skin preparation of Exidine 4% (chlorhexidine gluconate, CareFusion, Vernon Hills, IL) to the head. Ten hours elapsed between application of the surgical prep solution and the end of the surgery. After waking from successful resection of the meningioma, the patient reported eye pain ipsilateral to the operative left side. Upon bedside ophthalmic examination, the patient’s left eye was noted to have a large inferior hemicorneal epithelial defect with a well demarcated horizontal superior border corresponding to the upper eyelid margin. This was initially presumed to be a post-traumatic or exposure-related injury, and the patient was treated with erythromycin ophthalmic
ACCEPTED MANUSCRIPT
Discussion
TE D
M AN U
SC
RI PT
ointment applied to the affected eye. By postoperative day 13, there was complete resolution of the corneal abrasion but two months later the patient returned with complaints of ocular pain and decreased acuity on the operative side. Best corrected visual acuity was 20/50 in the affected left eye and 20/20 without correction in the fellow right eye. Intraocular pressure was normal in both the affected and fellow eyes, at 14 mm Hg and 13 mm Hg respectively (normal range, 10-21 mm Hg). Slit lamp examination showed sectoral microcystic corneal edema in the same distribution as the previous postoperative epithelial defect. The differential diagnosis included herpetic keratitis, so the patient was empirically treated with oral valacyclovir and a topical corticosteroid. However, after several weeks of treatment and lack of clinical response, the patient was sent to a corneal surgeon for evaluation. The corneal lesion’s interpalpebral location, sharply demarcated horizontal superior border, and depth to the corneal endothelium (Fig. 3) were all consistent with a toxic, penetrant substance gaining entry to the ocular surface via a partially open eyelid fissure but sparing the superior cornea where the eyelid was in apposition to the cornea. Endothelial toxicity was confirmed by direct imaging using specular microscopy, which demonstrated marked loss of endothelial cells (Fig. 4). Toxicity from the chlorhexidine-based surgical preparation was identified as the likely culprit given the onset of injury immediately after the mengioma resection. At present, the patient is being managed with conservative topical therapy but given extensive endothelial damage, the patient will likely need corneal transplantation in the future.
AC C
EP
Ocular injuries are infrequent following nonocular surgical procedures; one large retrospective study showed an incidence of about 0.056%.8 Anesthesiologists take various precautions including eyelid taping and bio-occlusive dressing (such as Tegaderm™) application to prevent ocular damage.9 Unlike the cases in this report, however, such ocular injuries are generally superficial, resulting in corneal epithelial disruption, and usually heal completely within 24-48 hours without specific therapy. The corneal epithelial defects in these cases took two and half months and 13 days, respectively, to heal and even after the corneal epithelial layer healed there was persistent damage to the deeper corneal layers, including the stroma and endothelium. In each of the above cases, the ocular damage caused by chlorhexidine was severe, causing protracted pain, visual loss lasting months, and a strong likelihood that corneal transplantation to restore good vision would be required. Recent data in the surgical literature has led to a shift toward using chlorhexidine rather than povidone-iodine in surgical prep because of reduced rates of surgical site infections associated with the former agent.1 While povidone-iodine at low concentrations is routinely and safely applied directly to the ocular surface for surgical prophylaxis in ophthalmic surgery, chlorhexidine has long been shown to cause considerable damage to ocular structures. 3 The injuries described range
ACCEPTED MANUSCRIPT
from epithelial defects that self resolve without sequelae, to irreversible opacification of the cornea requiring full-thickness corneal transplantation.
M AN U
SC
RI PT
The purpose of this report is to bring awareness to the potential ocular risks associated with chlorhexidine surgical prep, particularly for neurosurgical procedures. Surgical procedures calling for a prone position and those requiring preparation near the upper neck and face are at especially high risk for ocular injury resulting from inadvertent seepage of the chlorhexidine into the eyes.10 In both cases reported here, attempts were made to protect the eyes with a transparent adhesive nylon membrane (Tegaderm™), yet injury still occurred. We speculate that the porous, breathable nature of the transparent nylon dressing may have allowed the cationic chlorhexidine to freely permeate through the membrane into the underlying eye. It is also possible the occlusive nylon adhesive membrane did not remain fully adherent throughout the length of the procedure. Additionally, the duration of surgery may have played a role, since longer surgical cases have been associated with increased risk for ocular injury from inadvertent chlorhexidine exposure.8 The two patients in this report had surgical case lengths of seven hours and ten hours, respectively; that extended period of possible chlorhexidine exposure to the ocular surface could have contributed to the severe damage observed.
AC C
EP
TE D
The significant long-term patient consequences described in this report are reminders that special attention should be paid to protecting the eyes from exposure when employing chlorhexidine skin preparation in high-risk cases. Additionally, pre-operative discussion of surgical risks should include possible toxicity from surgical scrubs balanced against risk of surgical site infections. Measures that may lessen such risks include consideration of using povidone-iodine skin scrubs as an alternative to chlorhexidine in high-risk cases, avoiding applying the chlorhexidine preparation to the periocular skin, and placement of an absorbent eye pad over the closed eyes before applying the occlusive dressing, to serve as an additional layer of protection against seepage of chlorhexidine through or around the dressing.
ACCEPTED MANUSCRIPT
Darouiche RO, Wall MJJ, Itani KMF, et al. Chlorhexidine–Alcohol versus Povidone–Iodine for Surgical-Site Antisepsis. N Engl J Med. 2010;362(1):18-26. doi:10.1056/NEJMoa0810988.
2.
Lee I, Agarwal RK, Lee BY, Fishman NO, Umscheid CA. Systematic review and cost analysis comparing use of chlorhexidine with use of iodine for preoperative skin antisepsis to prevent surgical site infection. Infect Control Hosp Epidemiol. 2010;31(12):1219-1229. doi:10.1086/657134.
3.
Green K, Livingston V, Bowman K, Hull DS. CHlorhexidine effects on corneal epithelium and endothelium. Arch Ophthalmol. 1980;98(7):1273-1278. doi:10.1001/archopht.1980.01020040125020.
4.
Mac Rae SM, Brown B, Edelhauser HF. The corneal toxicity of presurgical skin antiseptics. Am J Ophthalmol. 1984;97(2):221-232.
5.
Phinney RB, Mondino BJ, Hofbauer JD, et al. Corneal Edema Related to Accidental Hibiclens Exposure. Am J Ophthalmol. 1988;106(2):210-215. doi:10.1016/0002-9394(88)90837-9.
6.
Hamed LM, Ellis FD, Boudreault G, Ii FMW, Helveston EM. Hibiclens Keratitis. Am J Ophthalmol. 1987;104(1):50-56. doi:10.1016/0002-9394(87)90292-3.
7.
Liu H-Y, Yeh P-T, Kuo K-T, Huang J-Y, Lin C-P, Hou Y-C. Toxic Keratopathy Following the Use of Alcohol-Containing Antiseptics in Nonocular Surgery. JAMA Ophthalmol. 2016;134(4):449. doi:10.1001/jamaophthalmol.2016.0001.
8.
Roth S, Thisted RA, Erickson JP, Black S, Schreider BD. Eye injuries after nonocular surgery. A study of 60,965 anesthetics from 1988 to 1992. Anesthesiology. 1996;85(5):1020-1027.
9.
Grixti A, Sadri M, Watts MT. Corneal Protection during General Anesthesia for Nonocular Surgery. Ocul Surf. 2013;11(2):109-118. doi:10.1016/j.jtos.2012.10.003.
EP
TE D
M AN U
SC
RI PT
1.
AC C
10. Yu H-D, Chou A-H, Yang M-W, Chang C-J. An Analysis of Perioperative Eye Injuries After Nonocular Surgery. Acta Anaesthesiol Taiwan. 2010;48(3):122129. doi:10.1016/S1875-4597(10)60043-4.
ACCEPTED MANUSCRIPT
SC
RI PT
Fig. 1. Case 1. Slit lamp photograph of the left cornea eight months after neurosurgical procedure showing significant opacification involving the visual axis.
M AN U
Fig. 2. Case 1. A: Specular microscopy of unaffected right eye showing normal endothelial cell density, 2427 cells/mm2. B: Specular microscopy of damaged left eye showing abnormal morphology and quantity (910 cells/mm2) of endothelial cells. These damaged endothelial cells do not regenerate.
EP
TE D
Fig. 3. Case 2. Slit lamp photograph of the affected left eye showing corneal haze.
AC C
Fig. 4. Case 2. A: Specular microscopy of unaffected right eye showing normal endothelial cell density, 2941 cells/mm2. B: Specular microscopy of damaged left eye showing abnormal morphology and quantity (518 cells/mm2) of endothelial cells. These damaged endothelial cells do not regenerate.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation Gregory Bever MD, Frank Brodie MD, David G. Hwang MD, FACS
•
Chlorhexidine 2% pre-surgical skin preparation used on the face in two neurosurgical cases led to permanent corneal damage Alternatives to Chlorhexidine include povidone-iodine 10% which does not have permanent corneal toxicity
AC C
EP
TE D
M AN U
SC
•
RI PT
Highlights
ACCEPTED MANUSCRIPT
Corneal Injury from Pre-surgical Chlorhexidine Skin Preparation Gregory Bever MD, Frank Brodie MD, David G. Hwang MD, FACS
RI PT
Abbreviations
AC C
EP
TE D
M AN U
SC
1) mm Hg: millimeters of mercury