Gonococcal Conjunctivitis in Adults: Case Report and Retrospective Review of Cases in Alberta, Canada, 2000–2016

Abstract

Background 

A case of gonococcal conjunctivitis (GC) prompted us to review the reported cases and treatment regimens of GC in Alberta, Canada.

Methods 

Gonococcal cases reported from 2000 to 2016 were extracted from the provincial sexually transmitted infection reporting database. The diagnosis of GC was based on a positive culture and/or nucleic acid amplification test from the eye.

Results 

A total of 45 cases of GC were reported in adults. Three quarters (75.6%; n = 34) of the cases were diagnosed using culture, 57.8% (n = 26) of cases were among men, and 55.5% (n = 25) were diagnosed since 2014. Very few (13.3%; n = 6) of the cases were treated according to current Canadian Guidelines on Sexually Transmitted Infections, using 2 g of ceftriaxone in combination with azithromycin or doxycycline. Results of test of cures were available for 15.6% (n = 7) of the cases and occurred within 10 to 79 days (median = 26 days) after treatment; all were negative.

Conclusions 

Gonococcal conjunctivitis was relatively uncommon in our region, but given its potential for severe manifestations and sequelae coupled with the rising rates of gonorrhea; it remains important to consider this diagnosis in sexually active individuals presenting with purulent conjunctivitis. Additional studies are needed to inform treatment recommendations and to evaluate outcomes of infection.

In Brief

Over the course of 16 years, 45 cases of gonococcal conjunctivitis were reported in adults. Very few cases were treated according to guidelines.

Gonococcal conjunctivitis (GC) is caused by the gram negative diplococcus Neisseria gonorrhoeae (NG) and is part of the spectrum of extra-genital gonococcal infections. Gonococcal conjunctivitis affects 2 main groups, neonates where this entity is called ophthalmia neonatorum, and sexually active persons.¹ Gonococcal ophthalmia neonatorum is acquired from an infected mother during delivery, and occurs in 30% to 50% of neonates exposed perinatally.² In non-neonates, infections result from autoinoculation or inoculation of infected genital secretions from a sexual partner.³ However, GC in both neonates and adults is relatively rare in the industrialized world.^1–3

The onset of GC is hyperacute and is characterized by chemosis and profuse purulent discharge.⁴ Symptoms can be rapidly progressive and ocular consequences can be devastating, due to the ability of NG to penetrate intact corneal epithelium.⁵ The clinical spectrum of this infection can vary greatly, with some cases presenting with isolated purulent conjunctivitis while others also involve the cornea. The degree of corneal involvement in gonococcal infection of the eye can also vary significantly, ranging from subepithelial and/or stromal infiltrates to corneal ulceration with subsequent thinning and perforation of the globe, resulting in endophthalmitis.⁴ Corneal involvement in these cases is of particular concern, because it can often lead to significant visual impairment. In the periorbital region, GC can cause significant lid edema which can mimic preseptal cellulitis.⁶ Occasionally, swelling may be severe enough to cause limitation of extraocular movements, which may lead to the misdiagnosis of orbital cellulitis. Because of the subtlety or lack of significant genitourinary (GU) symptoms, the diagnosis can often be delayed⁷; however, expeditious diagnosis and treatment is key in attempting to limit corneal involvement and, ultimately, in achieving preservation of vision.¹

Throughout the study period, the Canadian treatment guidelines for GC recommended ceftriaxone 2 g/d intravenous/intramuscular PLUS doxycycline/azithromycin while awaiting consultation.

The rate of gonorrhea in Alberta, Canada, has been increasing steadily from a low of 19/100,000 in 2000 to a peak of 87/100,000 in the midst of a provincial outbreak in 2016 (Fig. 1).⁸ Data on the frequency and outcomes of GC are limited. We sought to retrospectively evaluate the number and characteristics of reported cases, the treatment regimens, and outcomes of GC in Alberta, Canada, after a patient presented to our hospital with this condition.

Figure 1: Gonorrhea incidence rate per 100,000 population and rate of conjunctivitis per 100 gonorrhea cases (Alberta, 2000–2016).

CASE REPORT

A previously healthy 21-year-old woman presented to the emergency department (ED) with a 1-day history of left upper eyelid swelling, conjunctival injection, and profuse purulent discharge. Her visual acuity was unchanged, and she was discharged from the ED on oral cephalexin and topical erythromycin, tobramycin, and dexamethasone. An ocular swab was sent for culture and NG and Chlamydia trachomatis (CT) nucleic acid amplification testing (NAAT) using the GenProbe Aptima Combo 2 Assay (Hologic Inc, USA). Three days later, she re-presented with increasing ocular pain from her left eye. Urgent assessment by ophthalmology noted clinical findings of bacterial conjunctivitis. Her visual acuity was normal in her right eye (OD 20/20) but moderately decreased in her left eye (20/50). There was no significant corneal thinning or infiltrate. No systemic or GU symptoms were present. When the eye specimen NAAT result returned positive for NG the following day, the patient was admitted to hospital and treated with ceftriaxone 2 g intravenously (IV) single dose followed by 1 g IV daily for 4 additional days and azithromycin 2 g orally single dose and doxycycline 100 mg orally twice daily for 7 days; the latter was given as per ophthalmology protocol for the prevention of corneal melt.⁹ She also underwent frequent conjunctival lavage. Urine was also positive for NG NAAT. Human immunodeficiency virus (HIV) antibody was negative. The patient reported 3 male partners in the preceding 6 months with a new partner in the month before the current presentation.

Following discharge from the hospital, she was seen in follow up; her visual acuity had returned to baseline (20/20 in both eyes) and there were no signs of permanent ocular sequelae.

Retrospective Review of Alberta Cases

METHODS

Gonorrhea is a notifiable disease in both Canada and the province of Alberta. As part of the notifiable disease reporting guidelines, all gonorrhea cases are reported to the provincial Sexually Transmitted Infection (STI) Services. Testing providers complete a notification form which includes information on demographics, exposure risks and recent sexual partners. GC cases among adults (≥12 years) reported in Alberta from 2000 to 2016 were extracted from the provincial STI reporting database. The diagnosis of GC was based on a positive culture and/or NAAT result from the eye.

The data extract included date of diagnosis, sex, age, ethnicity, sexual partners, STI and HIV co-infection, testing provider, treatment, and additional anatomical sites that were infected. Testing provider was categorized as acute care, community provider, or a provincial STI clinic. For culture-positive isolates, minimum inhibitory concentrations (MIC) were provided for cefixime, ceftriaxone, ciprofloxacin, azithromycin, penicillin, and tetracycline using E-test conducted by the Alberta Provincial Laboratory of Public Health.

The molecular genotyping of the N. gonorrhoeae isolates using the Neisseria gonorrhoeae multi-antigen sequence typing (NG-MAST) method incorporated the amplification of the porin gene (por) and the transferrin-binding protein gene (tbpB) as previously described ¹⁰ The resulting polymerase chain reaction products were purified (QIA Quick PCR Purification Kit, Qiagen, Mississauga, Ontario or Agencourt AMPure, Beckman Coulter, Beverly, MA), and the DNA sequences of both strands obtained by the DNA Analyzer 3730xl (Applied Biosystems, Foster City, CA) were edited, assembled, and compared using software from DNAStar, Inc. (Madison, WI). The resulting sequences were submitted to the NG-MAST Web site (http://www.ng-mast.net/) to determine the STs.

Treatment data was divided into 5 categories based on the treatment containing at least: ceftriaxone 2 g IV, ceftriaxone 250 mg to 1 g IV, cefixime 800 mg orally, ciprofloxacin 500 mg orally, and other. Provincial treatment guidelines recommended all patients with extra-genital sites should have a test of cure (TOC). TOC with culture may be collected 3 days posttreatment and those collected using molecular testing 2–3 weeks after treatment. Sexually transmitted infection clinic charts were reviewed by an infectious disease physician to complete missing variables and retrieve information on TOC.

Descriptive analysis of GC patients by sex were compared using Chi-square or Fisher’s exact for discrete variables and Mann-Whitney for age, using IBM SPSS Statistics version 19.0 (IBM, Armonk, NY, USA). Ethics approval was obtained from the University of Alberta’s Health Research Ethics Board.

RESULTS

Forty-five GC cases were reported between 2000 and 2016. Over one half (55.5%; n = 25) of the cases were reported between 2014 and 2016. The rate of GC cases has ranged between 0 and 0.3 cases per 100 NG cases with the proportion of cases increasing with incidence rate increases (Fig. 2). One quarter of cases (26.7%; n = 12) also had GU infections.

Figure 2: A case of GC with profuse purulent discharge and chemosis.

Nearly 60% (n = 26) of cases were among men; however, there were no significant differences between sexes for ethnicity, geographic area of diagnosis, sexual partner, test type, HIV status, minimum treatment, or test of cure (Table 1). Women were younger at the time of diagnosis (median age, 21 years vs. 31 years; P = 0.03) and more likely to be diagnosed in an acute care facility (3.7% vs. 42.3%; P = 0.04) than men.

TABLE 1: Description of Gonococcal Conjunctivitis Cases by Sex (Alberta, 2000–2016; N = 45)

The majority (75.6%; n = 34) of cases were diagnosed using culture, with the majority (88.2%; n = 30) being culture-only. All isolates with available MIC data were susceptible to azithromycin (n = 30, <2 mg/L), cefixime (n = 30, <0.5 mg/L), and ceftriaxone (n = 28, <0.5 mg/L). A small number of isolates (16.7%; n = 5) were ciprofloxacin-resistant (≥1.0 mg/L), penicillin-resistant (13.3%; n = 4; ≥2.0 mg/L), and tetracycline-resistant (16.7%; n = 5; ≥2.0 mg/L). The NG-MAST sequence typing was available for 5 isolates collected between 2014 and 2015, and all were unique and previously reported in the province (ST-5, resistant to ciprofloxacin, penicillin, and tetracycline; ST-3935, tetracycline resistant; ST-5985, tetracycline resistant; ST-10451, resistant to ciprofloxacin, penicillin, erythromycin, and tetracycline; and ST-12122, resistant to penicillin, erythromycin, and tetracycline).

Treatment information was available for 93.3% (n = 42) of the cases. Of the cases where treatment date was available (n = 40), 55.0% (n = 22) were treated on the same day as the first specimen was collected. The remaining cases (n = 18) were treated within 1 to 30 days. Of the 42 cases with treatment records, 8 (19.0%) of the cases were treated with a regime containing ceftriaxone 2 g between 2008 and 2015, and only 6 (14.3%) of these cases met Canadian treatment guidelines with concomitant use of azithromycin or doxycycline. Nearly one half (47.6%; n = 20) of the cases were treated with a treatment containing ceftriaxone between 125 mg and 1 g, with use beginning in 2004 and continuing into 2016. Five (11.9%) cases were treated with a regime containing 500 mg of ciprofloxacin, all were treated between 2003 and 2007. Four (9.5%) cases were treated with a regime containing cefixime 800 mg; 2 cases in both 2015 and 2016. The remaining 5 (11.9%) cases were treated with other drugs, throughout the period. There was no significant difference between drug regimen and provider type (P = 0.81).

Results of TOC were available for 15.6% (n = 7) of the cases and occurred within 10 to 79 days (median, 26 days; IQR, 14–47 days) after treatment, all were negative. We excluded 2 cases in which TOC was only done at 102 and 229 days posttreatment. Cases with a TOC occurred between 2008 and 2016. The majority (71.4%; n = 5) of TOC cases were diagnosed at an acute care facility, with an additional case diagnosed each at a community providers and a STI Clinic. TOC cases were treated at baseline with ceftriaxone, 2 g (42.9%; n = 3); ceftriaxone, less than 2 g (42.9%; n = 3); and cefixime, 800 mg (14.3%; n = 1).

DISCUSSION

Our case report and case series highlight the importance of GC and the need for a high index of suspicion, especially when a patient presents with hyperacute conjunctivitis with significant chemosis and purulence (Fig. 2 shows the right eye of a patient with a similar presentation to our patient as a photograph of our patient was not available). Unfortunately, cases of GC may present with less typical findings, such as lid edema, which may mimic periorbital cellulitis, and is more likely to result in delays in diagnosis.⁶ The diagnosis may be further delayed by the subtlety or absence of significant GU symptoms.⁷

In our case report, despite the delay in diagnosis and treatment, the final outcome was excellent with normalization of visual acuity and no evidence of ocular abnormalities after appropriate treatment. Our decision to prolong treatment with ceftriaxone was based on the potential for corneal involvement and uncertainty of adherence to close follow-up and may have contributed to the successful outcome because gonococcal keratoconjunctivitis can progress rapidly to corneal destruction, endophthalmitis, and visual loss.⁴

Very few studies have reported on the outcomes of cases of GC,^11,12 our study provides one of the largest retrospective reviews in the medical literature. More than half of cases in our study were reported between 2014 and 2016, which may in part reflect the preferred use of highly sensitive NAAT tests to screen for gonorrhea instead of culture since 2014. In addition, reported gonorrhea rates have risen significantly in Alberta during the study period from 2015 to 2016. However, the fact that nearly 70% of the cases were diagnosed by culture contradicts NAAT as the main driver for the increase in GC cases in recent years.

Although the recommended treatment for gonococcal infection of the eye has not changed, the treatment for urogenital gonorrhea in Canada has changed over time as follows: (i) 2000 to 2006, ciprofloxacin 500 mg orally single dose; (ii) 2006 to 2011, cefixime 400 mg orally as a single dose; and (iii) 2011 to 2018, ceftriaxone in addition to azithromycin 1 g orally in a single dose.^13–15 Overall, of the 42 cases with treatment data available, only a small proportion of cases received guideline-concordant treatment with less than 20% treated with the recommended dose of ceftriaxone (2 g IV/IM) and only 14% having received it in combination with azithromycin or doxycycline, as recommended by the Canadian STI guidelines. Dual therapy is intended to potentially reduce the emergence and spread of NG resistance to cephalosporins and to potentially enhance treatment efficacy as supported by recent evidence.^13,16 Dual therapy in GC is particularly important considering that tissue penetration may constitute another barrier to successful treatment.

Expeditious diagnosis and treatment are key in attempting to limit corneal involvement and preserve vision.¹ Across the globe, variable recommendations for first-line therapy exist and include intravenous or intramuscular ceftriaxone with or without azithromycin^13,17–21 (Table 2). Treatment duration varies from 1 to 3 days, but advice is not provided about altering treatment dose or duration based on the severity of eye involvement.^13,17–21 European,¹⁹ United Kingdom¹⁸ and New Zealand¹⁷ guidelines support more prolonged courses of ceftriaxone given the relatively avascular nature of the cornea.

TABLE 2: Global Guidelines for the Treatment of Gonococcal Conjunctivitis in Adults/Adolescents

In our case series, most cases were diagnosed in acute care facilities followed by community-based practices and a small percentage in STI clinics. Despite the guideline recommendations to collect a TOC in extragenital sites, only 15.6% of our patients had follow-up NG testing done. Although patient compliance could have played a role, clinician knowledge of our current guidelines is likely suboptimal. These findings highlight the need to enhance awareness of treatment and TOC recommendations by clinicians outside of STI clinics, where most cases will likely be seen.¹¹ Some of the variation in the prescribed regimen may be related to the lack of consistency across global guidelines (Table 2).

Although GC without concurrent GU infection has been reported,^22,23 the approximately one quarter of cases in our sample with concurrent NG in the GU tract is probably an underestimate, as we were unable to confirm if all individuals were screened from the genital tract.

Our study is not without limitations. Its retrospective nature and small size provided limited statistical power to compare treatment regimens. As well, clinical indications for choosing treatment regimens were not available, and therefore, we are not able to rule out any bias this may have introduced. Treatment data are limited to records submitted to public health, and it is possible other treatments were received by the patient after reporting occurred. We did not have access to testing data to determine the proportion of patients with GC that were tested from other anatomical sites; it would have been relevant to know which patients were tested from anogenital and pharyngeal sites as GC most commonly results from auto-inoculation. Finally, information on visual outcomes, particularly degree of ocular involvement, need for surgery, and post-GC visual acuity, is not required by our reporting program and was not available for review.

Although GC has become relatively uncommon in the industrialized world, the rising rates of gonorrhea in many regions of the world make it an important consideration in sexually-active individuals presenting with purulent conjunctivitis. Current treatment duration recommendations for GC are based on very limited evidence, and this is reflected in the considerable variation in recommendations across international guidelines.^13,17–21 Also, the spectrum of GC severity has not been addressed in the guidelines as most data comes from case series and reports.^{1,6,7,23–30} Our study has identified the need for further studies of treatment and outcomes of GC, especially to determine if higher or longer doses of antibiotics are required with more severe disease presentations.