Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Apr 1.
Published in final edited form as: Cornea. 2022 Apr 1;41(4):484–490. doi: 10.1097/ICO.0000000000002893

Opioid Prescribing Patterns for Ulcerative Keratitis

Alexa R Thibodeau 1, Leslie M Niziol 1, Bradford L Tannen 1, Dena Ballouz 1, Maria A Woodward 1,2
PMCID: PMC8897229  NIHMSID: NIHMS1769032  PMID: 34620771

Abstract

Purpose:

To characterize rates of opioid prescription for different ulcerative keratitis types.

Methods:

This cohort study included patients diagnosed with ulcerative keratitis according to University of Michigan electronic health record data between September 1, 2014 and December 22, 2020. Ulcerative keratitis was categorized by etiologic type (bacterial, fungal, viral, acanthamoeba, inflammatory, polymicrobial, or unspecified) using rule-based data classification that accounted for billing diagnosis code, antimicrobial or anti-inflammatory medications prescribed, laboratory results and manual chart review. Opioid prescriptions were converted to milligram morphine equivalents (MME) and summed over 90 days from diagnosis. Opioid prescription rate and amount were compared between ulcerative keratitis types.

Results:

Of 3,322 patients with ulcerative keratitis, 173 (5.2%) were prescribed at least one opioid for pain management within 90 days of diagnosis. More patients with acanthamoeba (32.4%), fungal (21.1%), and polymicrobial (25.0%) keratitis were treated with opioids compared to bacterial (6.7%), unspecified (2.9%) or viral (1.8%) keratitis (all Bonferroni adjusted p<0.05). For the 173 patients that were prescribed opioids, a total of 353 prescriptions were given within 90 days of diagnosis, with half given within the first week after diagnosis. The quantity of opioid prescribed within 90 days from diagnosis was not significantly different between ulcerative keratitis types (p=0.6559). MME units prescribed ranged from 97.5 for acanthamoeba keratitis to 112.5 for fungal keratitis.

Conclusion:

Type of ulcerative keratitis may influence opioid prescription rate. Providers can better serve patients needing opioids for pain management through improved characterization of pain and development of more tailored pain management regimens.

Keywords: Ulcerative keratitis, Opioid, Prescribing

Introduction

Ulcerative keratitis is a vision-threatening ocular condition that is defined by the presence of a corneal stromal infiltrate. The annual incidence of ulcerative keratitis is estimated to be between 30,000 to 75,000 in the United States and 1.5 to 2 million worldwide.1–4 With approximately 7,000 nociceptors per square millimeter, the cornea is the most richly innervated tissue in the human body and a powerful pain generator.5–7 Pain control can be challenging for ulcerative keratitis given the high number of nociceptors, the multiple inflammatory mechanisms involved, the ocular surface toxicity of treatment medications, and a potentially prolonged clinical course.

Current methods of pain control for diseases of the ocular surface are limited. Topical anesthetics are fraught with adverse effects, including delayed epithelial healing, increased tear film instability, and decreased recoverability of organisms in corneal culturing. Overuse of topical anesthetics is a well-recognized potential harm and can cause serious morbidity from permanent tissue and nerve damage.8–14 As a result, ophthalmologists often use systemic opioid analgesics when pain is intolerable.

Declared a national public health emergency in 2017, the U.S. opioid epidemic is responsible for over 33,000 deaths per year, with nearly half of these deaths attributable to prescription opioids.15 Worldwide, physicians are becoming more aware of the dangers of opioids for patients introduced to them in the acute or perioperative setting. While adequate pain control is critical, surgeons now recognize the potential for misuse and long-term harms of opioids.16–27

In ophthalmology, most attention has focused on prescription of opioids following incisional ophthalmic surgeries. While best practices are becoming available for perioperative opioid prescribing, little is known about opioid prescribing in the non-perioperative setting.28,29 In this study, we explore opioid prescribing patterns for ulcerative keratitis. We seek to investigate opioid prescription rates and amounts by ulcerative keratitis type. Understanding patterns of opioid prescribing for ulcerative keratitis is the first step in developing recommendations and guidelines to improve the safety of prescribing and, in turn, help protect our patients.

Methods

Data Source

Data were obtained from the University of Michigan electronic health record (EHR) through the DataDirect self-service tool.30 Specifically, encounters with a billing diagnosis code for ulcerative keratitis (Supplemental Text 1, Supplemental Digital Content 1) that occurred within an outpatient or emergency setting between September 1, 2014 and December 22, 2020 were identified. Encounters where diagnosis codes mapped to a condition description of corneal erosion, herpes simplex conjunctivitis, peripheral ulcerative keratitis, or vitamin A deficiency with corneal ulceration were excluded. For the remaining encounters, data extracted included patient demographics (age, gender, race, ethnicity), diagnoses, lab orders, culture and pathology results, and prescription information for ophthalmic antimicrobial and opioid medications, including strength and quantity.

Ulcerative Keratitis Classification

Due to the sometimes-evolving nature of ulcerative keratitis diagnosis within a patient over time, a combination of manual chart review and rule-based data classification methods were used to determine the final diagnosis of a patient. Data within 90 days of initial diagnosis were reviewed to classify ulcerative keratitis type as bacterial, fungal, viral, acanthamoeba, inflammatory, polymicrobial, or unspecified, using the following system:

  1. Each encounter of a patient during the 90-day period was categorized for keratitis type based on the condition description associated with the billing diagnosis code (Supplemental Text 2, Supplemental Digital Content 2).

  2. Patients with only one ulcerative keratitis diagnosis type during the 90-day period were assigned that type as their final diagnosis.

  3. Patients with an unspecified diagnosis combined with one specific diagnosis (bacterial, fungal, viral, acanthamoeba, inflammatory) were assigned the more specific diagnosis.

  4. Patients with 2 or more specific ulcerative keratitis diagnoses underwent chart review to determine their final diagnosis.

  5. Patients who only had unspecified ulcerative keratitis diagnoses during the 90-day period were further investigated to determine a final diagnosis by assessing their prescribed medications.
    1. Patients prescribed a single medication class (antimicrobial, antifungal, antiviral, anti-inflammatory) were assigned the ulcerative keratitis diagnosis type associated with that class.
    2. Patients prescribed a single medication class in combination with an anti-inflammatory medication were assigned the more specific ulcerative keratitis diagnosis type associated with that medication class.
    3. Patients prescribed multiple medication classes underwent chart review to determine their final diagnosis

  6. Patients who only had unspecified ulcerative keratitis diagnoses during the 90-day period and no informative prescribed medications were further investigated to determine a final diagnosis by assessing their lab results, including culture or pathology results. Due to the limited lab data available, large number of cultures that return no growth, and the complexity of results, all lab data for these patients were chart reviewed to determine final ulcerative keratitis diagnosis

  7. Patients who only had unspecified ulcerative keratitis diagnoses during the 90-day period and no informative prescribed medication or lab results, retained their final diagnosis as unspecified.

Opioid Prescriptions

All medications prescribed in the 90-days from initial ulcerative keratitis diagnosis were reviewed for opioids (Supplemental Text 3, Supplemental Digital Content 3). Those opioid prescriptions originating in the ophthalmology department within 90 days of initial diagnosis or those originating in the emergency department (ED) within 7 days of initial diagnosis were retained. ED prescriptions were limited to 7 days to obtain opioids that were most likely prescribed for ulcerative keratitis and not another condition. Opioid prescriptions were converted to morphine milligram equivalent (MME) units and summed over the first 90 days since diagnosis.31 For interpretability, MME units were also converted to number of 5mg hydrocodone tablets. The primary outcome for analysis was two-fold: 1) The probability of receiving any opioid prescription within 90 days of initial ulcerative keratitis diagnosis, and 2) The quantity of opioid prescribed within 90 days of initial ulcerative keratitis diagnosis (MME units). The association between the primary outcome and ulcerative keratitis type was investigated.

Statistical Methods

Patient demographics of the sample are reported with descriptive statistics including means and standard deviations (SD) for continuous measures, and frequencies and percentages for categorical measures. Continuous demographic measures were compared between ulcerative keratitis types (bacterial, fungal, viral, acanthamoeba, inflammatory, polymicrobial, and unspecified) using analysis of variance (ANOVA) with Tukey adjusted post-hoc pairwise comparisons for continuous measures. Chi-square tests and Fisher exact tests were used to compare categorical demographic measures between ulcerative keratitis types, with significant tests followed by pairwise comparisons with Bonferroni adjustment. Descriptive statistics (mean, SD, median, interquartile range [IQR], frequency, percentage) were used to summarize opioid prescribing characteristics by ulcerative keratitis type and displayed visually with histograms and boxplots. The proportion of patients receiving at least one opioid prescription was compared between ulcerative keratitis types using a Fisher exact test followed by pairwise comparisons with Bonferroni adjustment. In the subset of patients with at least one opioid prescription, the MME units prescribed over 90 days was compared between ulcerative keratitis types using a Kruskal-Wallis test. Logistic regression was used to test for differences in the probability of receiving an opioid prescription over 90 days from diagnosis between ulcerative keratitis types while adjusting for patient demographic characteristics. Results are reported with odds ratios (OR) and 95% confidence intervals (CI). In the subset of patients who received at least one opioid prescription, linear regression was used to test for differences in MME units prescribed over 90 days since diagnosis between ulcerative keratitis types while adjusting for patient demographic characteristics. Due to the skewed distribution of MME units prescribed over 90 days, these data were modeled with a log-transformation. Results are reported with estimates (exponentiated to convert back to original scale) and 95% CIs. SAS version 9.4 (SAS Institute, Cary, NC) was used for statistical analysis.

Results

We identified a total of 3,322 patients with ulcerative keratitis. Patients were, on average, 52.1 years old at the time of their diagnosis (SD = 21.2, range = 1 month to 99 years), 57.7% female, 85.8% White (7.9% Black, 3.7% Asian, 2.6% Other), and 2.8% Hispanic (Table 1). The ulcerative keratitis type was bacterial in 47.2% of cases (n=1567), viral in 26.7% (n=886), fungal in 2.1% (n=71), acanthamoeba in 1.1% (n=37), inflammatory in 0.8% (n=28), polymicrobial in 0.4% (n=12), and unspecified in 21.7% (n=721). Demographic data stratified by ulcerative keratitis type are shown in Table 1. Age of patients showed significant differences with respect to ulcerative keratitis type (p<0.0001). Patients with bacterial keratitis were significantly younger (mean±SD, 49.0±21.2 years) than those with polymicrobial (67.6±14.0 years), viral (56.5±21.0 years), or keratitis of unspecified origin (53.5±20.6 years). Patients with acanthamoeba keratitis were significantly younger (43.2±20.8 years) than those with polymicrobial and viral keratitis (all p<0.05 after Tukey adjustment). A total of 170 minor patients (<18 years old) were diagnosed with ulcerative keratitis. The distribution of minor patients did not significantly differ between ulcerative keratitis type (p=0.15). Additionally, differences in ethnicity were noted between ulcerative keratitis types where significantly more patients with bacterial keratitis were of Hispanic ethnicity compared to those with unspecified keratitis (4.1% versus 1.4%; p=0.0044). There were no significant differences with respect to gender (p=0.8857) and race (p=0.1230) between ulcerative keratitis types.

Table 1.

Demographic characteristics of the ulcer sample, overall and stratified by ulcerative keratitis type

Continuous Variable Overall (n=3322) Acanthamoeba (n=37) Bacterial (n=1567) Fungal (n=71) Inflammatory (n=28) Polymicrobial (n=12) Unspecified (n=721) Viral (n=886) P-value**
(mean±SD) (mean±SD)
Age 52.1 ± 21.2 43.2 ± 20.8 49.0 ± 21.2 53.6 ± 17.3 52.7 ± 19.9 67.6 ± 14.0 53.5 ± 20.6 56.5 ± 21.0 <0.0001
Categorical Variable # (%) # (%) P-value**
Age <18 years 170 (5.1) 4 (10.8) 78 (5.0) 3 (4.2) 2 (7.1) 0 (0.0) 27 (3.7) 56 (6.3) 0.1490
Gender
 Female 1916 (57.7) 24 (64.9) 897 (57.2) 39 (54.9) 17 (60.7) 6 (50.0) 427 (59.2) 506 (57.1) 0.8857
 Male 1406 (42.3) 13 (35.1) 670 (42.8) 32 (45.1) 11 (39.3) 6 (50.0) 294 (40.8) 380 (42.9)
Race
 White 2790 (85.8) 35 (97.2) 1278 (83.3) 62 (89.9) 22 (85.6) 11 (100.0) 613 (86.3) 769 (88.8) 0.0839
 Black 256 (7.9) 1 (2.8) 145 (9.5) 2 (4.4) 1 (3.9) 0 (0.0) 55 (7.8) 51 (5.9)
 Asian 120 (3.7) 0 (0.0) 66 (4.3) 1 (1.5) 1 (3.9) 0 (0.0) 25 (3.5) 27 (3.1)
 Other 86 (2.6) 0 (0.0) 45 (2.9) 3 (4.4) 2 (7.7) 0 (0.0) 17 (2.4) 19 (2.2)
Ethnicity
 Hispanic 87 (2.8) 0 (0.0) 60 (4.1) 2 (2.9) 1 (3.9) 0 (0.0) 9 (1.4) 15 (1.8) 0.0051
 Non-Hispanic 2983 (97.2) 33 (100.0) 1390 (95.9) 68 (97.1) 25 (96.2) 10 (100.0) 653 (98.6) 804 (98.2)
Open in a new tab

SD, Standard Deviation

*

ANOVA,

**

Chi-Square (Gender) or Fisher exact (Age, Race, and Ethnicity) test

note: missing data includes n=70 for race, and n=252 for ethnicity

A minority of patients with ulcerative keratitis (5.2%, n=173 of 3322) receive at least one opioid prescription for pain management within 90 days of first diagnosis. There were significant differences in the percentage of patients prescribed opioids between ulcerative keratitis types (p<0.0001; Table 2). More patients with acanthamoeba (32.4%, n=12 of 37), fungal (21.1%, n=15 of 71), polymicrobial (25.0%, n=3 of 12), and bacterial (6.7%, n=105 of 1567) keratitis were treated with opioids compared to unspecified (2.9%, n=21 of 729) or viral (1.8%, n=16 of 886) keratitis (all Bonferroni adjusted p<0.05). A significantly lower percentage of patients with bacterial keratitis (6.7%) were treated with at least some opioids compared to acanthamoeba (32.4%) or fungal infections (21.1%) (all Bonferroni adjusted p<0.05).

Table 2.

Descriptive statistics for quantity of opioid prescribed within 90 days from ulcerative keratitis diagnosis, stratified ulcerative keratitis type

≥1 Opioid Rx MME units for those with ≥1 Opioid Rx 5mg Hydrocodone Tablets equivalent
Ulcer Type N # (%) Mean (SD) Median (IQR) Mean (SD) Median (IQR)
Acanthamoeba 37 12 (32.4) 242 (272) 97.5 (64.5, 350.0) 48.4 (54.3) 19.5 (12.9, 70.0)
Bacterial 1567 105 (6.7) 203 (280) 100.0 (60.0, 200.0) 40.5 (56.0) 20.0 (12.0, 40.0)
Fungal 71 15 (21.1) 330 (373) 112.5 (67.5, 550.0) 66.1 (74.6) 22.5 (13.5, 110.0)
Inflammatory 28 1 (3.6) 100 (.) 100.0 (100.0, 100.0) 20.0 (.) 20.0 (20.0, 20.0)
Polymicrobial 12 3 (25.0) 279 (408) 45.0 (40.5, 750.0) 55.7 (81.7) 9.0 (8.1, 150.0)
Unspecified 721 21 (2.9) 121 (88) 100.0 (75.0, 150.0) 24.2 (17.5) 20.0 (15.0, 30.0)
Viral 886 16 (1.8) 161 (137) 100.0 (70.0, 235.0) 32.3 (27.5) 20.0 (14.0, 47.0)
p<0.0001a p=0.6559b
Open in a new tab

SD, Standard Deviation; IQR, Interquartile Range

a

Fisher exact test;

b

Kruskal-Wallis (excluding inflammatory and polymicrobial categories due to small sample size)

The probability of being prescribed an opioid was significantly different between ulcerative keratitis types (p<0.0001; Table 3). Models were adjusted for gender, race, ethnicity and age (minor patients were less likely to receive an opioid prescription than older patients [1.8% versus 5.4%, respectively; p=0.0329]). All 21 comparisons between pairs of ulcerative keratitis types were investigated, and 12 showed significantly different probability of being treated with opioids in the first 90 days after diagnosis. Acanthamoeba (OR=14.95, p<0.0001), bacterial (OR=2.62, p=0.0002), fungal (OR=10.01, p<0.0001), and polymicrobial (OR=17.26, p=0.0001) keratitis types showed significantly higher odds of opioid treatment than unspecified keratitis. Acanthamoeba (OR=21.04, p<0.0001), bacterial (OR=3.69, p<0.0001), fungal (OR=14.10, p<0.0001), and polymicrobial (OR=24.29, p<0.0001) keratitis types also showed significantly higher odds of opioid treatment than viral keratitis. Acanthamoeba (OR=5.71, p<0.0001), fungal (OR=3.82, p<0.0001), and polymicrobial (OR=6.59, p=0.0088) keratitis types showed significantly higher odds of opioid treatment than bacterial keratitis. Lastly, acanthamoeba keratitis showed significantly higher odds of opioid treatment than inflammatory keratitis (OR=9.61, p=0.0381).

Table 3.

Regression model results for differences between ulcerative keratitis types with respect to probability of being treated with an opioid (logistic regression) and logarithm of 90 day MME units (linear regression). Model results are adjusted for age, gender, race, and education.

Logistic regression: ≥1 Opioid Rx Linear Regression: Log(90 day MME)
Comparison OR 95% CI P-value e(Estimate) 95% CI P-value
Acanthamoeba vs Unspecified 14.95 6.16, 36.25 <.0001 1.76 0.84, 3.69 0.1351
Bacterial vs Unspecified 2.62 1.57, 4.38 0.0002 1.22 0.75, 1.96 0.4211
Fungal vs Unspecified 10.01 4.77, 21.02 <.0001 1.73 0.91, 3.32 0.0959
Inflammatory vs Unspecified 1.55 0.20, 12.16 0.6743
Polymicrobial vs Unspecified 17.26 3.97, 74.93 0.0001
Viral vs Unspecified 0.71 0.36, 1.41 0.3257 1.09 0.58, 2.06 0.7884
Acanthamoeba vs Viral 21.04 8.53, 51.89 <.0001 1.61 0.75, 3.45 0.2157
Bacterial vs Viral 3.69 2.15, 6.32 <.0001 1.11 0.68, 1.84 0.6681
Fungal vs Viral 14.10 6.60, 30.11 <.0001 1.59 0.81, 3.11 0.1727
Inflammatory vs Viral 2.19 0.28, 17.23 0.457
Polymicrobial vs Viral 24.29 5.56, 106.22 <.0001
Acanthamoeba vs Bacterial 5.71 2.62, 12.41 <.0001 1.45 0.77, 2.71 0.2464
Fungal vs Bacterial 3.82 2.07, 7.05 <.0001 1.43 0.85, 2.40 0.1788
Inflammatory vs Bacterial 0.59 0.08, 4.45 0.6117
Polymicrobial vs Bacterial 6.59 1.61, 27.00 0.0088
Acanthamoeba vs Inflammatory 9.61 1.13, 81.63 0.0381
Fungal vs Inflammatory 6.44 0.80, 51.76 0.0797
Polymicrobial vs Inflammatory 11.10 0.97, 127.34 0.0532
Acanthamoeba vs Polymicrobial 0.87 0.18, 4.24 0.8591
Fungal vs Polymicrobial 0.58 0.13, 2.62 0.4786
Acanthamoeba vs Fungal 1.49 0.58, 3.84 0.4065 1.01 0.48, 2.16 0.9713
Open in a new tab

OR, Odds Ratio; CI, Confidence Interval; MME, Morphine Milligram Equivalent

Note: Logistic regression model excluded inflammatory (n=1) and polymicrobial (n=3) ulcerative keratitis types due to small sample size of patients who were prescribed at least one opioid; Exponentiated estimates from the logistic regression model are interpreted as the ratio of geometric means between ulcerative keratitis types.

For the 173 patients that were prescribed opioids, a total of 353 prescriptions were given within 90 days of diagnosis. Most patients received a single opioid prescription (n=102, 59.0%), followed by 14.5% (n=25) who received two, 11.6% (n=20) who received three, and 15.0% (n=26) who received four or more prescriptions. Figure 1 displays the distribution of timing of opioid prescriptions after initial diagnosis, by ulcerative keratitis type. Half of opioid prescriptions (50.4%, n=178) were given within the first week after diagnosis, including 18.1% (n=64) on the initial day of diagnosis. Most opioid prescriptions originated in the ophthalmology department (90.1%, n=318) with the remaining coming from ED and urgent care departments (9.9%, n=35). The median quantity of opioid prescribed within 90 days from diagnosis was 100 MME units (IQR=60–217, mean=203, SD=266) for the 173 patients. This is equivalent to a median of 20 5mg hydrocodone tablets (IQR=12–43, mean=41, SD=53).

Figure 1.

Figure 1.

Open in a new tab

Histogram displaying the distribution of opioid prescriptions over 90 days from diagnosis

The quantity of opioid prescribed within 90 days from diagnosis was not significantly different between ulcerative keratitis types (p=0.6559; Table 2; Figure 2). Specifically, median MME units prescribed ranged from 97.5 for acanthamoeba keratitis to 112.5 for fungal keratitis. In terms of number of 5mg hydrocodone tablets, median prescribed tablets ranged from 19.5 for acanthamoeba keratitis to 22.5 for fungal keratitis. These ranges exclude the inflammatory and polymicrobial keratitis types due to small sample size of patients who received at least one opioid prescription (n=1 and n=3, respectively). After adjusting for patient age, gender, race, and ethnicity, there were no significant differences observed in quantity of opioid prescribed within 90 days of diagnosis between ulcerative keratitis types (p=0.3589; Table 3).

Figure 2.

Figure 2.

Open in a new tab

Boxplots showing the distribution of opioid quantity prescribed over 90 days from ulcer diagnosis (MME units) in the subset of patients who had at least one opioid prescription (n=173), stratified by ulcerative keratitis type.

Discussion

In our study of 3,322 patients with ulcerative keratitis, a minority of patients (5.2%, n=173) received an opioid prescription within 90 days of diagnosis. The majority of patients who did receive opioids were provided a single prescription within 90 days of diagnosis and more than half of all prescriptions were written within the first week. Only a small subset of patients (15%, n=26) were provided 4 or more opioid prescriptions in the first 90 days. We found that the opioid prescription rate was higher for patients with acanthamoeba, fungal, and polymicrobial keratitis compared to inflammatory, viral or unspecified keratitis, however the median quantity of morphine milligram equivalents did not differ between keratitis types.

The pain associated with ulcerative keratitis can be severe and, thus, pain management is an essential component of any treatment plan. It is interesting and somewhat surprising that a small minority of patients received an opioid prescription within 90 days of diagnosis, and that half of all scripts were written within one week of diagnosis. The overall small percentage of patients who received opioids and the limited number of prescriptions per patient may suggest that ophthalmologists are cautious prescribers of opioids in the setting of ulcerative keratitis. Alternatively, it may be the case that ophthalmologists are not adequately assessing pain control, particularly later in the course of ulcerative keratitis, or have transitioned the responsibility of management to another provider, such as the primary care physician. Physicians have been shown to inadequately treat acute pain due to lack of training in the diagnosis and treatment of pain.32–35 In a recent survey, less than half of physician respondents agreed that their medical school curriculum had prepared them sufficiently to manage pain.36,37 Further, ophthalmologists specifically are likely to underestimate patient perceptions of pain.38,39

Differences in the degree of pain and treatment regimen associated with each ulcerative keratitis type may account for the differences observed in opioid prescription rates. Acanthamoeba and fungal keratitis were associated with the highest opioid prescription rate (32.4% and 21.1%, respectively). Severe ocular pain that is disproportionate to the degree of keratitis is a well-recognized hallmark of acanthamoeba keratitis and the presenting symptom in 91% of patients.40 Pain in acanthamoeba keratitis is related to perineural infiltration, which may be found in up to 63% of cases at 6 weeks.41,42 Topical biguanides and antifungals are particularly toxic to the corneal epithelium, while frequent drop administration and a long duration of therapy are typical.43 Earlier studies cite a mean duration of therapy of 70 days to 6 months for acanthamoeba keratitis.44,45 Meanwhile, the primarily fungistatic activity of currently available antifungal agents often necessitates a prolonged course of therapy, with a reported mean duration of 10 weeks.46,47 Exposure to multiple topical agents of different classes could explain the higher opioid prescription rate among keratitis in the polymicrobial category; investigation for an association between topical drop burden and opioid prescription and opioid refill rates is a potential area of future research. Conversely, viral keratitis was associated with the lowest opioid prescriptions rate in our study (1.8%). Herpetic viral infection is the most common cause of corneal hypoesthesia.48–50 Loss of corneal sensation in addition to preferred usage of oral rather than topical antiviral therapy may contribute to the decreased opioid prescription rate observed in patients diagnosed with viral keratitis. Of note, our study showed no apparent difference in the quantity of morphine milligram equivalents prescribed for different keratitis types. This observation suggests a tendency of ophthalmologists to prescribe a standard amount of opioids regardless of the degree of pain or keratitis type. Ophthalmologists may better serve patients through improved characterization of pain and development of more tailored pain management regimens.

Opioid prescribing practices by clinicians across all medical subspecialties has become a priority with the rise in opioid-related deaths over the past two decades. In ophthalmology, prescribing patterns have been analyzed largely in the context of perioperative pain management and at a global specialty level.28,29,51–54 Based on analysis of Medicare Part D Prescription Drug Program data from 2013 to 2015, Patel and Sternberg found that almost 90% of ophthalmologists wrote 10 or fewer opioid prescriptions annually, with a mean supply of 5 days.54 The median 100 MME (mean=203 MME) prescribed to keratitis patients within 90 days of diagnosis in this study is surprisingly higher than the average 80 MME that is prescribed within 90 days postoperatively for orbital and oculoplastic surgery.52 A query of a large national US insurer’s claims database found that 1.9% of incisional ocular surgeries were associated with an opioid prescription. In this same study, the opioid prescription rate was 6.1% for cornea surgeries alone.28 A recent cohort study of patients who underwent cornea surgery concluded that patients used fewer opioid tablets when surgeons prescribed fewer tablets yet continued to have adequate pain control. Interestingly, between 70–90% of these patients received an opioid prescription. This large percentage of patients prescribed opioids may be explained by the large portion of surface ablative procedures included in the study, as photorefractive keratectomy, superficial keratectomy and epithelium-off collagen cross-linking accounted for nearly 85% of all cases.51 Cornea specialists should be aware of the potentially higher analgesia requirements that their patients have.

There are some limitations to the study. First, data was from a single academic institution with a large referral base. Thus, delayed time to diagnosis or increased case complexity may have influenced opioid prescribing patterns and may not be reflective of other practice environments. Second, a large number of patients had unspecified keratitis despite a rigorous classification system. It is possible that this unspecified category includes milder or nearly resolved episodes of keratitis less likely to require opioids. This category was retained to speak to the prevalence of opioids across categories and also because many keratitis cases encountered in clinical practice do not have an identifiable etiology. Third, acanthamoeba keratitis was rare in this sample, so the significant higher odds of opioid treatment compared to unspecified, viral, bacterial and inflammatory keratitis may be an effect of small sample size of this rarer form of keratitis. Fourth, risk factors that potentially increase opioid use are not well captured within the electronic health record and could not be explored as predictors of opioid use for the entire 3,322-person sample. Our finding that patients with bacterial keratitis were significantly younger could be related to increased contact lens wear and poorer contact lens hygiene among younger patients in general.55–57 Future studies could investigate specific ophthalmic risk factors and other characteristics of keratitis, such as degree of pain, duration of symptoms and ulcer size, which may be associated with increased opioid use. Fifth, opioid data represents prescribed opioids, but those medications were not necessarily filled or used by the patient. Capture of opioid utilization data and patient reported outcomes to assess adequacy of pain control are future directions of research. Our results may serve as a foundation for such studies and for the development of guidelines for opioid prescribing in ulcerative keratitis. Development of an opioid prescribing protocol for keratitis could help guide ophthalmologists and primary care clinicians who initially manage and treat patients with acutely painful ulcerative keratitis.

Supplementary Material

Supplemental 1

Supplemental Text 1, Supplemental Digital Content 1: Billing diagnosis codes used to identify encounters with ulcerative keratitis in the electronic health record

Supplemental 3

Supplemental Text 3, Supplemental Digital Content 3: Opioid prescriptions pulled from the electronic health record

Supplemental 2

Supplemental Text 2, Supplemental Digital Content 2: Categorization of ulcerative keratitis types from condition descriptions associated with billing diagnosis codes

Acknowledgement:

Personnel effort (LMN) for this research at the University of Michigan was supported by generous donation of Ms. Susan Lane. The authors acknowledge the University of Michigan Medical School Research Data Warehouse and DataDirect for providing data, management, and distribution services in support of the research reported in this publication.

Financial Support:

National Eye Institute (R01EY031033 [MAW]), Research to Prevent Blindness, Career Advancement Award (MAW). The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Footnotes

Conflict of Interest: The authors have no proprietary or commercial interest in any of the materials discussed in this article

Disclosures: Unrelated to submitted work. Vortex Surgical (MAW, equity), Aviceda Ophthalmics (MAW, equity).

References

  • 1.Pepose JS, Wilhelmus KR. Divergent approaches to the management of corneal ulcers. Am J Ophthalmol. 1992;114:630–632. [DOI] [PubMed] [Google Scholar]
  • 2.Byrd LB, Martin N. Corneal Ulcer. In: StatPearls [Internet]. Stat Pearls Publishing, 2020. [PubMed] [Google Scholar]
  • 3.Whitcher JP, Srinivasan M, Upadhyay MP. Corneal blindness: a global perspective. Bull World Health Organ. 2001;79:214–221. [PMC free article] [PubMed] [Google Scholar]
  • 4.Keratitis Sharma S.. Biosci Rep. 2001;21:419–444. [DOI] [PubMed] [Google Scholar]
  • 5.Beuerman RW, Tanelian DL. Corneal pain evoked by thermal stimulation. Pain. 1979;7:1–14. [DOI] [PubMed] [Google Scholar]
  • 6.Goyal S, Hamrah P. Understanding Neuropathic Corneal Pain--Gaps and Current Therapeutic Approaches. Semin Ophthalmol. 2016;31:59–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Müller LJ, Marfurt CF, Kruse F, et al. Corneal nerves: structure, contents and function. Exp Eye Res. 2003;76:521–542. [DOI] [PubMed] [Google Scholar]
  • 8.Yagci A, Bozkurt B, Egrilmez S, et al. Topical anesthetic abuse keratopathy: a commonly overlooked health care problem. Cornea. 2011;30:571–575. [DOI] [PubMed] [Google Scholar]
  • 9.Lee MD, Driver TH, Seitzman GD. Cornea Specialists Do Not Recommend Routine Usage of Topical Anesthetics for Corneal Abrasions. Ann Emerg Med. 2019;74:463–466. [DOI] [PubMed] [Google Scholar]
  • 10.Varga JH, Rubinfeld RS, Wolf TC, et al. Topical anesthetic abuse ring keratitis: report of four cases. Cornea. 1997;16:424–429. [PubMed] [Google Scholar]
  • 11.Erdem E, Undar IH, Esen E, et al. Topical anesthetic eye drops abuse: are we aware of the danger? Cutan Ocul Toxicol. 2013;32:189–193. [DOI] [PubMed] [Google Scholar]
  • 12.Rosenwasser GO. Complications of topical ocular anesthetics. Int Ophthalmol Clin. 1989;29:153–158. [DOI] [PubMed] [Google Scholar]
  • 13.Pharmakakis NM, Katsimpris JM, Melachrinou MP, et al. Corneal complications following abuse of topical anesthetics. Eur J Ophthalmol. 2002;12:373–378. [DOI] [PubMed] [Google Scholar]
  • 14.Rocha G, Brunette I, Le François M. Severe toxic keratopathy secondary to topical anesthetic abuse. Can J Ophthalmol. 1995;30:198–202. [PubMed] [Google Scholar]
  • 15.Soelberg CD, Brown RE Jr., Du Vivier D, et al. The US Opioid Crisis: Current Federal and State Legal Issues. Anesth Analg. 2017;125:1675–1681. [DOI] [PubMed] [Google Scholar]
  • 16.Ung C, Yonekawa Y, Waljee JF, et al. Persistent Opioid Use After Ophthalmic Surgery in Opioid-Naive Patients and Associated Risk Factors. Ophthalmology. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Southren DL, Moser S, Abu-Amara H, et al. Opioid Fills for Lumbar Facet Radiofrequency Ablation Associated with New Persistent Opioid Use. Anesthesiology. 2020;132:1165–1174. [DOI] [PubMed] [Google Scholar]
  • 18.Bennett KG, Kelley BP, Vick AD, et al. Persistent Opioid Use and High-Risk Prescribing in Body Contouring Patients. Plast Reconstr Surg. 2019;143:87–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Habermann EB. Are Opioids Overprescribed Following Elective Surgery? Adv Surg. 2018;52:247–256. [DOI] [PubMed] [Google Scholar]
  • 20.Hah JM, Bateman BT, Ratliff J, et al. Chronic Opioid Use After Surgery: Implications for Perioperative Management in the Face of the Opioid Epidemic. Anesth Analg. 2017;125:1733–1740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Clarke H, Soneji N, Ko DT, et al. Rates and risk factors for prolonged opioid use after major surgery: population based cohort study. BMJ. 2014;348:g1251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Lawal OD, Gold J, Murthy A, et al. Rate and Risk Factors Associated With Prolonged Opioid Use After Surgery: A Systematic Review and Meta-analysis. JAMA Netw Open. 2020;3:e207367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Harbaugh CM, Nalliah RP, Hu HM, et al. Persistent Opioid Use After Wisdom Tooth Extraction. JAMA. 2018;320:504–506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zaveri S, Nobel TB, Khetan P, et al. Risk of Chronic Opioid Use in Opioid-Naïve and Non-Naïve Patients after Ambulatory Surgery. J Gastrointest Surg. 2020;24:688–694. [DOI] [PubMed] [Google Scholar]
  • 25.Shah AS, Blackwell RH, Kuo PC, et al. Rates and Risk Factors for Opioid Dependence and Overdose after Urological Surgery. J Urol. 2017;198:1130–1136. [DOI] [PubMed] [Google Scholar]
  • 26.Crystal DT, Blankensteijn LL, Ibrahim AMS, et al. Quantifying the Crisis: Opioid-Related Adverse Events in Outpatient Ambulatory Plastic Surgery. Plast Reconstr Surg. 2020;145:687–695. [DOI] [PubMed] [Google Scholar]
  • 27.Levy N, Quinlan J, El-Boghdadly K, et al. An international multidisciplinary consensus statement on the prevention of opioid-related harm in adult surgical patients. Anaesthesia. 2021;76:520–536. [DOI] [PubMed] [Google Scholar]
  • 28.Kolomeyer AM, Yu Y, VanderBeek BL. Association of Opioids With Incisional Ocular Surgery. JAMA Ophthalmol. 2019;137:1283–1291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Charlson ES, Feng PW, Bui A, et al. Opioid Prescribing Patterns Among American Society of Ophthalmic Plastic and Reconstructive Surgery Members in the Medicare Part D Database. Ophthalmic Plast Reconstr Surg. 2019;35:360–364. [DOI] [PubMed] [Google Scholar]
  • 30.DataDirect: a self-serve tool for data retrieval. 2015. Available at the University of Michigan at: https://datadirect.med.umich.edu/.
  • 31.CDC compilation of benzodiazepines, muscle relaxants, stimulants, zolpidem, and opioid analgesics with oral morphine milligram equivalent conversion factors. Available at: https://www.cdc.gov/drugoverdose/resources/data.html. Accessed May 1, 2021.
  • 32.Sinatra R. Causes and consequences of inadequate management of acute pain. Pain Med. 2010;11:1859–1871. [DOI] [PubMed] [Google Scholar]
  • 33.Bertakis KD, Azari R, Callahan EJ. Patient pain in primary care: factors that influence physician diagnosis. Ann Fam Med. 2004;2:224–230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Phillips DM. JCAHO pain management standards are unveiled. Joint Commission on Accreditation of Healthcare Organizations. Jama. 2000;284:428–429. [DOI] [PubMed] [Google Scholar]
  • 35.Silvoniemi M, Vasankari T, Vahlberg T, et al. Physicians’ self-assessment of cancer pain treatment skills--more training required. Support Care Cancer. 2012;20:2747–2753. [DOI] [PubMed] [Google Scholar]
  • 36.Garcia J, Ohanisian L, Sidley A, et al. Resident Knowledge and Perception of Pain Management. Cureus. 2019;11:e6107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Pizzo PA, Clark NM. Alleviating suffering 101--pain relief in the United States. N Engl J Med. 2012;366:197–199. [DOI] [PubMed] [Google Scholar]
  • 38.Chang SH DU, Gupta PK, Hovanesian JA, et al. Consider patient perception of pain when planning cataract procedures. Available at: https://www.healio.com/news/ophthalmology/20171102/consider-patient-perception-of-pain-when-planning-cataract-procedures. Accessed May 1, 2021.
  • 39.Jiang L, Zhang K, He W, Zhu X, Zhou P, Lu Y. Perceived Pain during Cataract Surgery with Topical Anesthesia: A Comparison between First-Eye and Second-Eye Surgery. J Ophthalmol. 2015;2015:383456. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Chynn EW, Lopez MA, Pavan-Langston D, et al. Acanthamoeba keratitis. Contact lens and noncontact lens characteristics. Ophthalmology. 1995;102:1369–1373. [DOI] [PubMed] [Google Scholar]
  • 41.de Lacerda AG, Lira M. Acanthamoeba keratitis: a review of biology, pathophysiology and epidemiology. Ophthalmic Physiol Opt. 2021;41:116–135. [DOI] [PubMed] [Google Scholar]
  • 42.Dart JK, Saw VP, Kilvington S. Acanthamoeba keratitis: diagnosis and treatment update 2009. Am J Ophthalmol. 2009;148:487–499.e482. [DOI] [PubMed] [Google Scholar]
  • 43.Maycock NJ, Jayaswal R. Update on Acanthamoeba Keratitis: Diagnosis, Treatment, and Outcomes. Cornea. 2016;35:713–720. [DOI] [PubMed] [Google Scholar]
  • 44.Sun X, Zhang Y, Li R, et al. Acanthamoeba keratitis: clinical characteristics and management. Ophthalmology. 2006;113:412–416. [DOI] [PubMed] [Google Scholar]
  • 45.Radford CF, Lehmann OJ, Dart JK. Acanthamoeba keratitis: multicentre survey in England 1992–6. National Acanthamoeba Keratitis Study Group. Br J Ophthalmol. 1998;82:1387–1392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Srinivasan M Fungal keratitis. Curr Opin Ophthalmol. 2004;15:321–327. [DOI] [PubMed] [Google Scholar]
  • 47.Bunya VY, Hammersmith KM, Rapuano CJ, et al. Topical and oral voriconazole in the treatment of fungal keratitis. Am J Ophthalmol. 2007;143:151–153. [DOI] [PubMed] [Google Scholar]
  • 48.Bonini S, Rama P, Olzi D, et al. Neurotrophic keratitis. Eye (Lond). 2003;17:989–995. [DOI] [PubMed] [Google Scholar]
  • 49.Davis EA, Dohlman CH. Neurotrophic keratitis. Int Ophthalmol Clin. 2001;41:1–11. [DOI] [PubMed] [Google Scholar]
  • 50.Liesegang TJ. Corneal complications from herpes zoster ophthalmicus. Ophthalmology. 1985;92:316–324. [DOI] [PubMed] [Google Scholar]
  • 51.Woodward MA, Zhang Y, Tannen B, et al. Association of Limiting Opioid Prescriptions With Use of Opioids After Corneal Surgery. JAMA Ophthalmol. 2019;138:76–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Xie Y, Joseph AW, Rudy SF, et al. Change in Postoperative Opioid Prescribing Patterns for Oculoplastic and Orbital Procedures Associated With State Opioid Legislation. JAMA Ophthalmol. 2021;139:157–162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Starr MR, Patel SV, Bartley GB, et al. Impact of Standardized Prescribing Guidelines on Postoperative Opioid Prescriptions after Ophthalmic Surgery. Ophthalmology. 2020;127:1454–1459. [DOI] [PubMed] [Google Scholar]
  • 54.Patel S, Sternberg P Jr., Association Between Opioid Prescribing Patterns and Abuse in Ophthalmology. JAMA Ophthalmol. 2017;135:1216–1220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Chalmers RL, Wagner H, Mitchell GL, et al. Age and other risk factors for corneal infiltrative and inflammatory events in young soft contact lens wearers from the Contact Lens Assessment in Youth (CLAY) study. Invest Ophthalmol Vis Sci. 2011;52:6690–6696. [DOI] [PubMed] [Google Scholar]
  • 56.Keay L, Edwards K, Naduvilath T, et al. Microbial keratitis predisposing factors and morbidity. Ophthalmology. 2006;113:109–116. [DOI] [PubMed] [Google Scholar]
  • 57.Bennett L, H YH, Tai S, et al. Contact Lens Versus Non-Contact Lens-Related Corneal Ulcers at an Academic Center. Eye Contact Lens. 2019;45:301–305. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental 1

Supplemental Text 1, Supplemental Digital Content 1: Billing diagnosis codes used to identify encounters with ulcerative keratitis in the electronic health record

NIHMS1769032-supplement-Supplemental_1.docx (12.5KB, docx)
Supplemental 3

Supplemental Text 3, Supplemental Digital Content 3: Opioid prescriptions pulled from the electronic health record

NIHMS1769032-supplement-Supplemental_3.docx (14.5KB, docx)
Supplemental 2

Supplemental Text 2, Supplemental Digital Content 2: Categorization of ulcerative keratitis types from condition descriptions associated with billing diagnosis codes

NIHMS1769032-supplement-Supplemental_2.docx (12.9KB, docx)

RESOURCES