Introduction

Infectious keratitis is a leading cause of blindness and hospitalisation worldwide, incurring £2,855 per admission [1]. Causative organisms demonstrate marked geographical and temporal variation, and there are limited epidemiological data necessary to establish disease trends [2]. We analysed infectious keratitis patterns and trends in a regional context, informing prevention and management strategies.

Methods

We audited corneal scrapes from a major UK NHS trust, serving Newcastle and Gateshead’s densely populated and deprived areas (population: 556,181) [3], from January 2012 to 2023. Sight-threatening corneal ulcers meeting local criteria underwent urgent investigation. Cultures were incubated for at least a week, longer for suspected Acanthamoeba. Microbial trends were analysed using age-adjusted Poisson regression. We considered only the first non-commensal microbial agent each admission (±30 days) to avoid duplications.

Results

We found 600 culture-positive cases involving 72 species and 31 genera; 10% were repeat cases. Bacteria comprised 88.8%, fungi 7.0%, and Acanthamoeba 4.2%. Positive scrape yield averaged 37.9% per admission. Median age was 59 (IQR 42–75) years, stable throughout.

The commonest genera were Staphylococcus (25%), Pseudomonas (22%), Streptococcus (12%), and Moraxella (10%). Gram-positive and negative representation was equal (44%, 43%). Although Staphylococcus cases remained stable (0% annual change, P = 0.99), Pseudomonas cases rose significantly (+7%, P = 0.003), establishing it as the predominant infectious agent over the past four years (Fig. 1). Pseudomonas keratitis peaked July to November (27–42%) vs. rest of year (5–20%). There were no consistent changes in other pathogens’ rates, including Acanthamoeba (−8%, P = 0.34; Supplementals).

Fig. 1: Area plot showing the trends of the most prevalent infectious keratitis pathogens over the past decade.
figure 1

Each line represents the cumulative proportion of culture-positive cases for that year. The relative incidence of Staphylococcus has remained stable, while Pseudomonas has shown an upward trend in recent years.

Full size image

Peak months were May, July, August, September, and October (9–10% each). The busiest period was the last week of June and the first week of July (Supplementals). Gram-positive pathogens prevailed in May (57%) and January (55%), while Gram-negative pathogens predominated in November (58%) and August (55%). On weekdays, most samples were processed Tuesdays (21%) and Saturdays (18%). These timings do not consider potential delays in sample handling.

Discussion

We reveal a sustained increase in Pseudomonas-related keratitis rates across Northern England, recently surpassing Staphylococcus cases (Fig. 1). This aligns with other UK regions’ trend of declining Gram-positive infections, favouring Gram-negative pathogens like Pseudomonas [4, 5]. This shift may reflect changing contact lens use behaviours, a recognised risk for Gram-negative infections. However, it remains unclear why this practice would not have led to a similar trend in Acanthamoeba keratitis – this disparity could stem from softer water compared to the South [6].

We also confirm bacteria as the main cause of infectious keratitis, comprising 89% of culture-positive cases (previously reported: 91–93%) [2]. Staphylococcus and Pseudomonas dominated our sample (25%, 22%), surpassing rates in other UK areas like neighbouring Sunderland (14%, 12%) [2]. Non-bacterial keratitis rates were comparable [2], hinting at limited microbial diversity possibly due to elevated antibiotic use [7]. Additionally, we confirmed increased culture-positive cases in warmer months [5], and identified increased rates on Tuesdays and Saturdays, likely due to higher patient visits to our Eye Casualty the previous day [3].

Further epidemiological study is urgently needed to establish UK-wide infectious keratitis trends and identify preventable causal relationships.