The Duration of Analgesic Drugs is the Major Determinant for Numbers Needed to Treat (NNT) Calculations

 

Statement of problem: The randomized controlled trial is one of the most powerful tools in analgesic drug research. The outcomes from one trial can be difficult to compare with results from other trials. The concept "numbers needed to treat" (NNT) gives the number of patients one needs to treat to achieve a predetermined analgesic effect in one patient. The analgesic effect is defined as at least a 50% reduction in pain intensity or at least a 50 % effect of pain relief. An analgesic drug effect can basically be separated into 2 basic elements. One element is the analgesic depth (i.e. to which extent the pain is gone) and the other is the analgesic duration (i.e. for how long will the pain be gone). NNT combines these 2 elements, but does not tell which element that contributes most to the value of the NNT. The NNT has also been criticized for being misleading.

Purpose: The purpose of the present trial was to investigate if the NNT can be manipulated by the choice of analgesic drugs and the observation times used for calculating NNTs.

Materials and Methods: In a randomized controlled parallel-group trial a total of 350 patients (200 females/150 males), ASA class I-II, with a mean age 24.5 years (range 18-30 years) and mean BMI 22.8 (SD 2.8) were divided into 7 treatment groups being treated with capsules of powdered tablets of ibuprofen 800 mg (IBU800), 600 mg (IBU600), 400 mg (IBU400), acetaminophen 500 mg (APAP500) or 1000 mg (APAP1000), acetaminophen 1000 mg + codeine 60 mg (APAPCOD), and placebo (P), respectively. The patients were subjected to removal of a single third molar (lower jaw) using a mean volume of 4.2 ml (SD 1.2 ml) lidocaine 2% + adrenalin 1:80 000. Mean onset time for local anaesthesia was 10:00 mm:ss (SD 02:00 mm:ss) and duration of surgery 20:00 mm:ss (SD 09:00 mm:ss). Pain intensity (PI) was subjectively assessed by patients on a numerical rating scale running from 0 to10. Informed patient consent and Norwegian Ethical Committee approval was obtained prior to trial.

Data analysis: The sum pain intensity difference score (SUMPID) was calculated for each treatment group. SUMPIDs were calculated following completion of surgery up to 3, 4, 5, and 6 hours for each treatment group. Corresponding NNTs were calculated (PASW Statistics ver. 18.0.1).

Results: Calculated NNTs for a 6 hour observation period were; IBU800 1.7, IBU600 1.9, IBU400 2.9, APAP1000 4.5, APAP500 7.1, APAPCOD 3.3, and P 25.0. NNTs for a 5 hour period were IBU800 1.6, IBU600 2.0, IBU400 2.8, APAP1000 4.5, APAP500 6.3, APAPCOD 2.9, and P 25.0. NNTs for a 4 hour period were IBU800 1.9, IBU600 2.1, IBU400 2.8, APAP1000 3.8, APAP500 3.5, APAPCOD 2.5, and P 25.0. NNTs for a 3 hour period were IBU800 2.4, IBU600 2.5, IBU400 3.3, APAP1000 3.8, APAP500 3.5, APAPCOD 2.5, and P 25.0.

Conclusion: This trial shows how NNTs can be manipulated. The manipulation being done by the choice of analgesic drug doses, type of drug and the observation time for analgesic effect. An increase of the ibuprofen dose while increasing the observation period will decrease the NNT. An increase of the dose of acetaminophen, with or without codeine, while increasing the observation period will increase the NNT. Clinicians should be critical when assessing effects of analgesic drugs based on NNT calculations.

References:

 

Moore A, McQuay H. Numbers needed to treat derived from meta-analysis. NNT is a tool, to be used appropriately. BMJ 1999;319:1200.

 

Smeeth L, Haines A, Ebrahim S. Numbers needed to treat derived from meta-analyses -- sometimes informative, usually misleading. BMJ 1999;318:1548-51.