Investigation into postoperative endophthalmitis and lessons learned

Area of Science:

• Ophthalmology
• Infectious Disease Epidemiology
• Biostatistics

Background:

• Endophthalmitis is a serious intraocular infection following ocular surgery.
• Determining whether sporadic cases represent a true outbreak is critical for public health interventions.
• Statistical methods are essential for evaluating the probability of disease clusters occurring by chance.

Purpose of the Study:

• To investigate a cluster of endophthalmitis cases to ascertain if it constituted a genuine outbreak.
• To evaluate the utility of statistical approaches in outbreak detection.
• To identify potential contributing factors to the observed endophthalmitis cases.

Main Methods:

• Utilized Bayesian statistical analysis to assess the probability of endophthalmitis cases occurring by chance.
• Reviewed potential factors associated with outbreaks of post-operative endophthalmitis.
• Case series investigation conducted at an outpatient facility in Sunderland, England.

Main Results:

• Bayesian statistical analysis proved effective in identifying the endophthalmitis outbreak.
• The primary factor identified was the surgeon's cessation of subconjunctival antibiotic prophylaxis.
• The decision to stop prophylaxis was based on a perceived lack of strong evidence for its efficacy.

Conclusions:

• Bayesian statistical analysis is a valuable tool for confirming the occurrence of outbreaks.
• Discontinuing established prophylactic treatments based solely on the absence of robust evidence can lead to adverse patient outcomes.
• Clinical decisions regarding prophylaxis should consider the potential for harm when evidence of benefit is lacking.