Paracetamol (acetaminophen) with codeine in postoperative pain

Clinical bottom line

Paracetamol with codeine is an effective analgesic. The NNT for at least 50% pain relief with a single dose of paracetamol 1000 mg plus codeine 60 mg was 2.2 (1.7 to 3.9) based on information on 197 patients. Paracetamol 600/650 mg plus codeine 60 mg had an NNT of 4.2 (3.4-5.3) for at least 50% pain relief over 4-6 hours in patients with moderate to severe pain compared with placebo based on information from 1136. Paracetamol plus codeine produced significantly more pain relief than paracetamol alone.

Paracetamol is an important non-opiate analgesic, accounting for over 5 million prescriptions in England alone (1995). In combination with codeine it accounts for a further 6.4 million prescriptions (1996). Paracetamol/codeine combinations are also widely available without prescription.

Systematic review

RA Moore, S Collins, D Carroll, HJ McQuay. Paracetamol with and without codeine in acute pain: a quantitative systematic review. Pain 1997 70:193-201.

RA Moore, SL Collins, D Carroll, HJ McQuay, J Edwards. Single dose paracetamol (acetaminophen), with and without codeine, for postoperative pain. The Cochrane Library, Update Software, Oxford 2000.

LA Smith, RA Moore, HJ McQuay, D Gavaghan. Using evidence from different sources: an example using paracetamol 1000 mg plus codeine 60 mg. BMC Medical Research Methodology 2001, 1:1 ( http://www.biomedcentral.com/1471-2288/1/1 ).

J Barden, JE Edwards, RA Moore, SL Collins, HJ McQuay. Single dose paracetamol (acetaminophen) plus codeine for postoperative pain. The Cochrane Library, Update Software, Oxford 2002.

Date review completed: 2002
Number of trials included: 25 paracetamol plus codeine vs. placebo / 13 paracetamol plus codeine vs. paracetamol
Number of patients: 1385 paracetamol+codeine vs. placebo / 794 paracetamol+codeine vs. paracetamol
Control groups: placebo and paracetamol
Main outcomes: 4-6 hr TOTPAR (predominantly); number-needed-to-treat (NNT) for 50% pain relief (with 95% confidence intervals); relative benefit (with 95% confidence intervals).

Inclusion criteria were full journal publication of trials of paracetamol and paracetamol plus codeine in acute postoperative pain; single oral dose; randomised; placebo-controlled; double-blind; moderate to severe baseline pain; adult populations; group sizes at least 10; sufficient data to calculate the area under the curve for pain relief (TOTPAR).

For each trial the mean TOTPAR values for paracetamol and placebo groups were converted to the percent of maximum total pain relief based on the categorical pain scales (%maxTOTPAR). These values were then converted to dichotomous information on the proportion, and then the number of patients, who achieved at least 50%maxTOTPAR. A number-needed-to-treat for at least 50% pain relief and the relative benefit of the treatment were then calculated.

Findings

Paracetamol plus codeine versus placebo

A single oral dose of paracetamol 600/650 mg plus codeine 60 mg generated an NNT of 4.2 (3.4-4.3) for at least 50% pain relief over 4-6 hours in patients with moderate to severe pain compared with placebo. Paracetamol 1000 mg plus codeine 60 mg (Figure 1; Table 1) had an NNT of 2.2 (1.7 to 3.9) compared with placebo.

Figure 1: Randomised comparisons of paracetamol 600/650 mg plus codeine 60 mg versus placebo

Table 1: NNTs for at least 50% pain relief over 4-6 hours for paracetamol plus codeine at different doses compared with placebo

		At least 50% pain relief number/total (%)
Paracetamol + Codeine dose (mg)	Number of trials	Paracetamol+ Codeine	Placebo	Relative benefit (95% CI)	NNT (95% CI)
300 + 30	4	56/215 (26)	14/164 (8)	3.0 (1.8 to 5.3)	5.7 (4.0 to 9.8)
500 + 30	1	13/49 (27)	7/45 (16)	0.7 (0.7 to 3.9)	9.1 (3.7 to 18.7)
600/650 + 60	17	261/636 (42)	83/487 (18)	2.4 (1.9 to 2.9)	4.2 (3.4 to 5.3)
800 + 60	1	16/44 (36)	0/21 (0)	15.6 (1.0 to 249)	2.7 (2.0 to 4.5)
1000 + 60	3	65/114 (57)	8/83 (9)	4.8 (2.6 to 8.8)	2.2 (1.7 to 2.9)

Higher doses of both paracetamol and codeine led to increased efficacy and lower NNTs (Figure 2). With paracetamol 600/650 mg plus codeine 60 mg 42% of patients with initial pain of moderate or severe intensity had at least 50% pain relief over 4-6 hours, as did 57% with paracetamol 1000 mg plus codeine 60 mg.

Figure 2: 95% confidence intervals of NNTs with paracetamol/codeine combinations

Paracetamol plus codeine versus paracetamol

Addition of 60 mg of codeine was associated with significant extra analgesic effect when compared directly with paracetamol alone. An additional 13%-22% of patients had at least 50% pain relief when codeine 60 mg was added to paracetamol (Figure 3; Table 2).

Figure 3: Randomised comparisons of paracetamol 600/650 mg plus codeine 60 mg versus paracetamol 650 mg

Table 2: NNTs for at least 50% pain relief over 4-6 hours for paracetamol plus codeine at different doses compared with paracetamol

		At least 50% pain relief number/total (%)
Paracetamol + Codeine dose (mg)	Number of trials	Paracetamol + Codeine	Paracetamol alone	Relative benefit (95% CI)	NNT (95% CI)
600/650 + 60	10	165/309 (54)	129/313 (41)	1.3 (1.1 to 1.5)	8.2 (5.0 to 22.7)
1000 + 60	3	74/109 (68)	52/108 (46)	1.4 (1.1 to 1.8)	5.1 (3.1 to 14.5)

Adverse effects

There were no serious adverse effects which necessitated withdrawal from any study. For paracetamol 600/650 mg there was significantly higher levels of drowsiness (NNH 11 (7-18)) and dizziness (NNH 19 (11-50)) with paracetamol plus codeine compared with placebo.

Comment

Paracetamol plus codeine is an effective analgesic combination, with low NNTs for at least half pain relief over 4-6 hours at doses of paracetamol 600/650 mg plus codeine 60 mg and paracetamol 1000 mg plus 60 mg. Though there was limited information for the combination of paracetamol 1000 mg plus codeine 60 mg, there was considerable supporting evidence from other combinations. In addition, three trials lacking a placebo had similar event rates for paracetamol 1000 mg plus codeine 60 mg as did those with a placebo. Six other trials with some design issues like the use of different pain measures that meant that they could not be combined showed the combination to be better than placebo or comparators.