Scientific and Technologycal Research

Héctor J. Meléndez*, José Rafael Ramos**

* Médico anestesiólogo, MSc, profesor asociado de postgrado de Anestesiología, Universidad Industrial de Santander, Bucaramanga, Colombia. melendez@uis.edu.co

** Médico anestesiólogo, Universidad Industrial de Santander, Bucaramanga, Colombia.

Recibido: febrero 13 de 2009. Enviado para modificaciones: enero 27 de 2010. Aceptado: marzo 15 de 2010.

Pain management is the main concern in 57 % of patients that undergo surgery (1). Pain can produce pulmonary, cardiac and gastrointestinal complications among the others, and lengthens hospital stay and increases costs.

Worldwide opioids and specifically morphine are the most common medications used for postoperative pain management, but their side effects, respiratory depression, nausea, vomiting, constipation and urinary retention have limited their use (2). In Colombia low availability of morphine, in addition of fearful poorly trained personnel in recovery rooms has minimized opioid use (3).

Currently, multimodal pain management combining opioids with non opioids has demonstrated a decrease in opioid requirement. However, trials with this type of therapy have been controversial in the magnitude of side effects reduction, although they have shown significant reductions in treatment costs (4).

In Colombia, acute postoperative pain management most commonly is done by general practitioners or surgeons which has led to suboptimal results and only when difficulties to manage appear, specialists are consulted.

The results of a pilot study performed by us, in which the combination of dipyrone (2.5 g) and meperidine (0.5 mg per kilogram) showed an effectiveness of 75 % with a decrease of 60 % in side effects like nausea and vomiting in addition to the prevalent use of dipyrone (5,6,7), with its low cost and the scenario of pain management described above; led us to develop this controlled clinical trial to confirm its efficacy to propose it as a good alternative with minimal side effects for pain management by non specialists.

A double-blind controlled clinical trial was performed. Patients with ages between 15 and 75 years with an ASA score of I-II scheduled for gynecological or general surgery laparotomy under regional or general anesthesia were included. Exclusion criteria were: existence of communication problems, hemodynamic instability, indwelling epidural catheters, contraindication for the medications studied, or history of addiction to opioids.

Pain was assessed with the visual analog scale which has been well validated in which acute postoperative pain is considered to be 4 and above (8).

The sample size was calculated based on a previous study in Colombia which showed an incidence of 37 % of acute postoperative pain (3) and using the bioequivalence hypothesis of proportions. The following statement was used:

With 0.5 mg per kilogram of meperidine in addition to 2.5 g of dipyrone the same efficacy in acute pain management is achieved than using 1 mg per kilogram of meperidine in addition to 2.5 g of the dipyrone; calculating for a type I error of 5 % (95 % significance) and a type II error of 10 % (90 % power); if a 50 % difference in the use of opioids to consider that the treatments are clinically equivalent and a true difference of 10 % between them, then 58 patients for group would be needed.

A randomized block allocation was performed to guarantee homogeneity in the group size. A random number table was generated and assigned centrally.

The research assistant prepared the medication using a concealed syringe with the same volume (20 ml) to ensure blinding, and delivered it to the anesthesiologist who was blinded. The blinded medication was to be administered when skin closure was begun. Afterwards the medication was prescribed according to protocol every 6 hours for 24 hours. The prepared syringes where delivered to the pain clinic nurse responsible of administering the medications and performing the follow up.

Before administering subsequent doses, pain intensity was assessed as well as the side effects of the medications. If the patient complained of pain before the scheduled dose, rescue analgesic was provided, using a dose of 0.25 mg per kg of meperidine to be repeated if pain persisted 15 minutes later. If pain was not controlled an assessment by the pain clinic staff was requested.

All patients received 4 mg of dexamethasone with 20 mg of metoclopramide or 4 mg of ondensatron to decrease the incidence of nausea and vomiting, a very common side effect of meperidine. The choice of these medications was subject to availability and their administration was recorded in the instrument designed for this purpose.

The data recorded in the instrument was analyzed in a Microsoft Excel 2007 table specifically coded. Quantitative variables were described with central trends and dispersion measurements; they were grouped and a frequency table was built. Qualitative variables were described as percentages with their confidence intervals; and quantitative variables were described using averages, medians, standard deviations and their respective confidence intervals (9).

Baseline demographics and clinical characteristics of the patients in the two groups were analyzed to verify if they were comparable and thereafter a normal distribution of each of them was done. Hypotheses tests were performed to assess the differences for both the continuous as well as dichotomous or qualitative variables.

The clinical characteristics of the patients are described using means and portions and their respective 95 % confidence interval (IC 95 %). Fisher’s exact test and Student’s T test with 0.05 significance were used to assess the differences between dichotomous and continuous variables respectively. The differences between more than two subgroups were tested with ANOVA. A two and multiple variable analysis was performed calculating the relative risk (RR) as an impact measurement. Statistically significant differences were considered as those in which the pvalue was below 0.05 (10,11). The analysis was done with STATA 9.0 statistical analysis software (12).

This study was done in the Hospital Universitario de Santander and was approved by the interinstitutional Hospital Universitario de Santander- Universidad Industrial de Santander ethics committee and had signed informed consent by all enrolled patients.

The 107 patients were distributed two groups (study group n = 50, control group n = 57) which was 91 % of the calculated sample. Five percent of the patients were lost for follow-up 12 hours after surgery. These losses were due to protocol violations, persistent vomiting or voluntary dismissal. The two groups did not show any statistically significant difference in baseline characteristics. Average age was 40 years, average weight was 62 kg and the mean body mass index was 26 (table 1).

Postoperative acute pain incidence at 6 hours was 28 % in the study group and 26.32 % in the control room. The relative risk was higher in the study group (RR = 1.06; CI95 % 0.57-1.98), without a difference between them (table 2).

Even though there were no significant differences in baseline characteristics (p = 0.119), the relationship of postoperative acute pain according to the surgery performed and the expected pain severity (moderate or severe) (13) was analyzed finding no significant differences (p = 0.477). The analysis according to the severity of expected pain (moderate or severe) and that assessed at 6 hours did not show any significant difference (p = 0871 and p = 0.714) (tables 3,4 and 5).

During the observed time, the absolute and relative risk of pain were higher in the first 18 hours in the study group, but did not reach statistically significant difference. The maximum pain measured in the first 6 hours was higher in the study group and in the next time intervals it was the same and at 24 hours, pain was absent in the study group (table 6).

The average time for rescue dose was 125 and 94 minutes in the study and control group respectively, without significant differences (p = 0.449). In the first 6 hours, the study group had a higher absolute risk of rescue dose requirement (26 % vs. 24.56 %) and relative risk (1.05) without significant differences (p =0.861) (table 7).

The control group had a significant difference of absolute risk of developing higher incidence of side effects (p = 0.0055). Dizziness was the side effect with higher incidence in the study group without significant difference. The other side effects had higher relative risks with statistically significant difference in the control group except respiratory depression (table 8).

It is important to highlight that most research on acute postoperative pain is done with patient controlled analgesia (PCA) which increases costs. However, the literature still reports high incidence of this type of pain with rates between 46 % and 53 % (14,15).

The results of our studies suggest that the addition of dipyrone to meperidine allows a decrease in 50 % in the need for opioids to control postoperative acute pain which is higher than that found in Paris by Aubrun et al (16) who combined paracetamol with morphine reporting a decrease in 37 %, and are better results than those found at hospital San Vicente de Paúl in Medellín were the most common medication was dipyrone with poor results in 69 % of the patients (6). The significant decrease in opioid use we found, allows us to promote the combined use for postoperative acute pain management.

Even though in 27 % of the patients pain control was not effective, it was higher than that reported by Camacho in 1995 (3). These results are still to unsatisfactory even when pain management improved in 30 %, we still have significant pain in one out of four patients and the goal is to achieve 0 to 3 VAS score of postoperative pain and that is where our research is focused.

The decrease in side effects with lower doses is well known and was corroborated in our study. The higher incidence of dizziness could be produced by the additive effect of dipyrone and meperidine at the central level without excluding the effects of anesthesia. Worldwide, literature reports up to 73 % of adverse effects with opioids (6) and other studies have shown controversial results in which benefits of multimodal analgesia are not found (14).

The incidence of 26.3 % and 28 % of postoperative acute pain in both groups (control and study group) without significant difference, suggests that pain control was not optimal as the incidence of acute pain was far from ideal.

It is important to consider that there may be different interpretations in the process of calculating the sample size according to the software used, so we can conclude that we may be exposed to commit a type II error due to the lack of power. For example, if we recalculate the sample size with the bioequivalence proportional parameters either one or two tailed taking as baseline a 50 % decrease in pain (from 37 % down to 18.5 %) with a maximum difference of 0.05, the sample size required would be 3283 and 9659 patients per group respectively (Pontificia Universidad Javeriana’s software) whereas if we calculate the sample size with the same parameters with EpiInfo, the sample size would only be 101 patients.

1. Pogatzki-Zahn E, Zahn PK, Brennan TJ. Postoperative pain-clinical implications of basic research. Best Pract Res Clin Anaesthesiol. 2007;21(1):3-13.

2. Buvanendran A, Reuben S, Kroin J. Recent advances in nonopioid analgesics for acute pain management. Tech Reg Anesth Pain Manag. 2007;11(1):19-26.

3. Camacho N. Incidentes en la unidad de recuperación posquirúrgica Hospital Universitario Ramón González Valencia (tesis). Bucaramanga: Universidad Industrial de Santander; 2006.

4. Elia N, Lysakowski C, Tramèr MR. Does multimodal analgesia with acetaminophen, nonsteroidal antiinflammatory drugs, or selective cyclooxygenase- 2 inhibitors and patient-controlled analgesia morphine offer advantages over morphine alone? Meta-analyses of randomized trials. Anesthesiology. 2005;103(6):1296-304.

5. McQuay H. Opioids in pain management. Lancet. 1999;353(9171):2229-32.

6. Cardona E, Castaño ML, Builes AM, Castro G. Manejo del dolor postquirúrgico en el Hospital Universitario San Vicente de Paúl de Medellín. Rev Colomb Anestesiol. 2003;31(2):111-7.

7. Arcila-Herrera H, Barragán-Padilla S, Barbolla-Escoboza J, Canto-Solís A, Castañeda-Hernández G, León-González M, et al. Consenso de un grupo de expertos mexicanos. Eficacia y seguridad del metamizol (dipirona). Gac Méd Méx. 2004;140(1):99-101.

8. Loeser J, Butler S, Chapman C, Turk D. Bonica, terapéutica del dolor. 3ª ed. México: Mc Graw-Hill Interamericana; 2003.

9. Fleiss JL. Determining sample sizes needed to detect a difference between two proportions. In: Fleiss JL, Levin BA, Cho Paik M. Statistical methods for rates proportions. 2nd ed. New York: John Wiley & Sons; 1981. p. 33-49.

10. Greenland S. Introduction to regression modeling. In: Rothman K, Greenland S, Lash T. Modern epidemiology. 2nd ed. Buenos Aires: Lippincott; 1998. p. 401-32.

11. Rothman K, Greenland S, Lash T. Precision and validity of studies. In: Modern epidemiology. 2nd ed. Philadelphia: Lippincott; 1998. p. 119-33.

12. StataCorp. Stata Stadistical Software: Release 9.0. College Station: Stata Corporation; 2006.

13. Puebla F. Tipos de dolor y escala terapéutica de la OMS. Dolor iatrogénico. Oncología. 2005;28(3):139- 43.

14. Bustos M, González E, Gallego J, Mayo M, Pallares J. Estudio comparativo de la eficacia analgésica del tramadol versus morfina-metamizol tras histerectomía abdominal. Rev Soc Esp Dolor. 2003;10(7):411-8.

15. Macintyre P. Safety and efficacy of patient controlledanalgesia. Br J Anaesth. 2001;87(1):36-46.

16. Aubrun F, Kalfon F, Mottet P, Bellanger A, Langeron O,Coriant P, et al. Adjunctive analgesia with intravenous propacetamol does not reduce morphine related adverse effects. Br J Anaesth. 2003;90(3):314-9.