The study was a prospective double masked, randomised controlled trial. We compared the effectiveness of paracetamol versus placebo as pre‐emptive analgesia in PRP in patients with diabetic retinopathy. The McGill pain questionnaire (MPQ) and visual analogue scales (VAS) were used to grade the pain.
Ethics committee approval was obtained and all participants gave informed consent.
Sequential patients with newly diagnosed proliferative diabetic retinopathy, attending medical retina clinics at Sunderland Eye Infirmary, were invited to participate in the study.
Exclusion criteria for the study were history of paracetamol sensitivity or allergy, severe liver disease, pregnancy, patients on regular analgesics, previous PRP, and age less than 18 years old.
Sixty patients were enrolled in the study. The patients were randomised to paracetamol or placebo via an electronic randomisation function (SPSS 12.1). The randomisation tables were sent to the pharmacist who allocated the patient a randomisation number and gave the patient the appropriate medication. The patients were instructed to take the medication every 6 hours, for 3 days, starting 1 day before the laser treatment. Non‐compliant patients were excluded from the study. The participants, nursing staff, and medical staff were masked to the type of trial medication the patient received.
Two consultant ophthalmologists with an interest in retinal disease performed the laser treatment.
The treatment was standardised as follows: PRP, total of 1200 burns, 400 μm spot size, duration of pulse 0.1 seconds, aiming for a moderately grey burn, avoiding the horizontal midlines, maximum treatment time of 30 minutes, mean treatment time 15 minute (as recommended by EDTRS).11
Benoxinate drops 0.4% were used for topical anaesthesia, before contact lens use.
Pain was assessed at two points following the laser treatment:
- Immediately (within 15 minutes) after the laser treatment to evaluate the pain experienced during laser
- Twenty four hours after the laser treatment to evaluate pain experienced following the laser treatment.
The pain was assessed using, at both time points, two techniques.
Firstly, an MPQ was used. The MPQ was designed to enable the patient to specify both the quality and degree of pain experienced.12
It is based on the observation that each type of pain is characterised by a distinctive constellation of words. It analyses four major aspects of pain: sensory (1–10), affective (11–15), evaluative (16), and miscellaneous (17–20) (fig 1). The pain rate index (PRI) is based on the rank values of the words. The word in each subclass implying the least pain is given a value of 1, the next word is given a value of 2, etc. The rank values of the words chosen by a patient are summed to obtain a score separately for the sensory, affective, evaluative, and miscellaneous words, in addition to providing a total score (subclass 1–20, which is PRI‐T).12
Present pain intensity (PPI) represents the number‐word combination chosen as the indicator of overall pain intensity at the time of administration of the questionnaire.12
Secondly, a VAS, from 0 to 10 was used.
Figure 1MPQ administered to the patients included in trial.
A nurse trained to guide the patient through the MPQ and VAS supervised the first assessment. The same nurse conducted a second pain assessment over the telephone, 24 hours after the first pain assessment.
The patients were provided with 30 mg codeine phosphate tablets to use in case of breakthrough pain.
The primary outcome of the study was the assessment of pain immediately after the laser treatment and at 24 hours after laser, as expressed by the total pain rate index (PRI‐T). The secondary outcomes were: MPQ subgroup results analysis and VAS results analysis immediately after laser and analysis of pain at 24 hours after laser as assessed by both MPQ and VAS.
Statistical power was set at 80% (5% significance level) in order to detect a difference of 30% in reported pain (PRI‐T) between paracetamol and placebo groups for the primary outcome.
The Mann‐Whitney test for two independent samples was used to test between group differences, while the Wilcoxon matched pairs test was used to measure change over time within groups. Frequency data were examined via Fisher's exact test for two proportions, with the addition of 95% (exact) confidence intervals where appropriate.