Abstract

Objective

The purpose of this study was to establish the streptozotocin (STZ)-induced diabetic model with rats and investigate the antinociceptive effect of combination of Tramadol (TR) and Acetaminophen (NAPA) on the animal model for the first time.

Methods

Diabetic model was induced by a single injection of STZ (60 mg/kg, intraperitoneal). Nociceptive thresholds were measured by means of electronic von Frey test, hot-plate test, and tail-flick test. On the 28th day of diabetes induction, diabetic rats with significant hyperalgesia were randomly divided into three groups: TR, NAPA, and TR-NAPA combination group. Each group was randomly divided into four subgroups. Three geometric series of drugs were given to each group respectively. Antinociceptive effects of the drugs were assessed at 15, 30, 60, 120, and 180 minutes after drug administration. 50% Maximum antinociceptive effect of each drug was determined by probit analysis, whereas interaction between TR and NAPA was evaluated by isobolographic analysis.

Results

Hyperalgesia, along with hyperglycemia, developed 4 days after STZ injection and persisted at all tested time points until 28 days. TR, NAPA, and TR-NAPA combination administration all produced dose-dependent antinociceptive effects. Isobolographic analysis showed a significant deviation of TR/NAPA 50% maximum antinociceptive effect (in tail-flick test, but not in von Frey test) from the additive line.

Conclusions

Combination of the two drugs produces an additive antinociceptive effect in tail-flick test, whereas probable additive antinociceptive effect in von Frey test in painful diabetic neuropathy rats.

Keywords

Tramadol; acetaminophen; drug interactions; diabetes mellitus; pain: neuropathic;

1. Introduction

Painful diabetic neuropathy (PDN) is generally considered to be one of the most common complications of diabetes mellitus (DM), and nearly 16–26% of diabetic patients experience chronic pain.¹ PDN is often associated with mood and sleep disturbances, and thus can impair the quality and expectancy of life.2, 3 The exact cause and mechanism of PDN remain unclear. Therefore, the treatment of diabetic neuropathic pain is largely aimed at relieving painful symptoms. However, conventional pain relievers often show to inefficacy and are complicated by intolerable side effects.⁴

Tramadol (TR) is a centrally acting, non-narcotic analgesic for use in treating mild to moderate pain. Evidence has shown that it is effective in treating the pain of diabetic neuropathy and improving patients’ quality of life.⁵ However, the treatment of severe pain with TR is unsatisfactory because of its low analgesic efficacy. And the dose-dependent adverse effects may also hamper its clinical use.⁶

The aim of increasing the analgesic effect and decreasing the adverse effect of TR led us to explore drug coadministration. Acetaminophen (NAPA) is the most commonly used pain reliever with minor side effects. The antinociceptive effects of TR (Group T) and NAPA (Group A) were studied, and additionally, their synergism was evaluated through isobolographic analysis.

2. Methods

The principles outlined in the Declaration of Helsinki were strictly followed throughout all experiments. Male adult Sprague-Dawley rats (body weight 200–230 g, Wei tong li hua, Beijing, China) were housed in groups in cages (n = 6 per cage) in a room with controlled temperature (21–22°C), and alternately light-dark cycle (12 h/12 h), with access to food and water ad libitum.

Diabetes was induced by an intraperitoneal injection of streptozotocin (STZ) at 60 mg/kg (Sigma Chemicals, St. Louis, MO, USA) in freshly prepared 0.1mM citrate buffer (pH 4.5) on the fourth day after animal arrival (allowing 3-day’s habituation to the environment) after an overnight fast. Control animals received an equivalent volume of vehicle. Confirmation of hyperglycemia was made 3 days later by measurement of the glucose concentration of the venous blood of the tail (Accu-Chek Advantage II, Ronche Diagnostic GmbH, Germany). Animals with blood glucose levels >15mM were considered diabetic (DM group, n = 99), and those whose levels were below 7.1mM were assigned to STZ control group (n = 16). (The initial total number of rats was 115.) In this study, care was taken during the induction of diabetes to avoid the induced illness to become too severe. All animals were observed daily and weighed regularly during the study period.

Diabetic rats were tested for hyperalgesia (by means of electronic von Frey test, hot-plate test, and Tail-immersion test) on the 4^th, 7^th, 14^th, 21^st and 28^th days after STZ injection. Diabetic rats with significant hyperalgesia were considered as PDN rats and used in drug evaluation study. To evaluate dose- and time-dependent antinociceptive effects of each drug, on the 28^th day of DM induction, the PDN rats (n = 99, of which 96 rats were divided into three groups, and the 3 remainders were discarded) were randomly divided into three groups, i.e. Group T, Group A, and Group T/A (n = 32 for each group). Each group was randomly divided into four subgroups (n = 8 each). A single injection of TR (1.5 mg/kg, 3 mg/kg, 6 mg/kg, 12 mg/kg for respective subgroup), NAPA (26 mg/kg, 52 mg/kg, 104 mg/kg, 208 mg/kg for respective subgroup) or TR/NAPA (1.5/13 mg/kg, 3/26 mg/kg, 6/52 mg/kg, 12/104 mg/kg for respective subgroup) was given intraperitoneally. Antinociceptive effects of the drugs were assessed at the 15, 30, 60, 120 and 180 minutes after drug administration by means of three behavioral tests. And nociceptive latency was expressed as % of the maximum possible effect, where % maximum possible effect = (postdrug threshold − predrug threshold) × 100/(cut-off time − predrug threshold).

Mechanical sensitivity was measured with a calibrated electronic von Frey pressure algometer (electronic von Frey anesthesiometer; IITC Inc., Life Science Instruments, Woodland Hills, CA, USA), which consisted of a hand-held force transducer fitted with a 0.7 mm² polypropylene tip. The electronic pressure-meter test was carried out in the way described by Vivancos et al.⁷with slight modification. Each studied rate was isolated in an acrylic cage (20 × 20 × 40 cm) with a wire grid floor (0.2 × 0.2 cm) 15–30 minutes before test to allow environmental adaptation. Before paw stimulation, the animals were quiet, without exploratory defecation or urination and resting on paws. The von Frey probe was manually applied to the plantar surface of the hindpaw with a gradually increasing pressure until a brisk withdrawal of the hindlimb was provoked. With the electronic pressure-meter, the intensity of the stimulus was automatically recorded when the paw was withdrawn. The equipment was calibrated to determine the pressure linearly until 80 g. The stimulation of the paw was repeated until the animal had brought out three approximate measurements (the difference between the highest and the lowest measurement should be less than 10 g). The average of the three measurements was the mechanical withdrawal threshold (MWT).

Tail-flick (warm water) test was carried out as below. The rat was immobilized by wrapping the body with a soft cloth in a proper strength. The tail was immersed in warm water (53 ± 0.3°C) contained in a bath for a length of 1.5 cm until tail withdrawal (flicking response) or signs of struggle was observed. To prevent tissue damage, a cut-off time of 10 seconds was set. The test was repeated four times at interval of 30 seconds for each rat. The average of the latter three values was taken as Tail-flick latency (TFL).

Hot-plate test was carried out as below. Animals were individually placed on a hot-plate (RCY-2, Yilian Medical Supply Inc., Shanghai, China) with the temperature adjusted to 53 ± 1°C. The latency to the first sign of hindpaw licking or jump response to avoid the heat was taken as an index of the pain threshold; the cut-off time was 10 seconds to avoid damage to the paw. Each rat was tested three times at 20-second interval. The average of the three measurements was taken as thermal withdrawal threshold (TWT).

2.1. Data analysis

Differences between responses were evaluated by one-way analysis of variance followed by Bonferronis t test. The differences were considered to be statistically significant at p < 0.05.

The dose of each drug required to produce 50% maximum antinociceptive effect (ED₅₀) was determined by probit analysis (SPSS v12.0, SPSS Inc., Chicago, Illinois, USA).

Interactions between TR and NAPA were assessed with the use of isobolograms (microsoft office excel 2003). In the present study, isobolograms were constructed by connecting the ED₅₀ of NAPA plotted on the abscissa with the ED₅₀ of TR plotted on the ordinate to obtain an additivity line. The additivity line contains the dose pairs that produce an ED₅₀ equal to the ED₅₀ of TR or NAPA alone. Dose pairs that fall below the additivity line suggest that an ED₅₀ was reached with lesser quantities of the drugs, indicating synergism. In contrast, experimental points representing dose pairs that fall above the line are suggestive of subadditivity.

3. Results

Four days after STZ injection, the diabetic rats had significantly higher blood glucose levels (19.65 ± 0.26 mmol/L) than the control rats (6.16 ± 0.18 mmol/L; p < 0.001). There was a marked decrease in the body weight of STZ-treated rats as compared with control rats (Fig. 1). The general health was monitored strictly. Although the diabetic rats had a significantly reduced body weight, yet all animals demonstrated normal behavior during the study period. Neither severe dehydration nor severe muscle weakness was observed.

Download full-size image

Fig. 1. Trends of body weight in diabetic group (n = 99), STZ control group (n = 16), and vehicle control group (n = 5). Data are expressed as means ± standard deviation. ∗p < 0.05 versus control groups. STZ, streptozotocin.

The nociceptive threshold was significantly lower in diabetic rats as compared with the basal value tested in both tail-flick and electronic von Frey assays. Hyperalgesia was evident on Day 4 and the maximum decrease in pain threshold was observed 4 weeks after STZ injection (Fig. 2, Fig. 3). Although the diabetic rats showed a significant decrease in TWT as compared with the basal value, the rats in two control groups seemed to show the similar trend and there were no significant differences among the groups (Fig. 4).

Download full-size image

Fig. 2. Trends of mechanical withdrawal threshold in diabetic group (n = 99), STZ control group (n = 16), and vehicle control group (n = 5). Data are expressed as means ± standard deviation. ∗p < 0.05 versus control groups. STZ, streptozotocin.

Download full-size image

Fig. 3. Trends of tail-flick test in diabetic group (n = 99), STZ control group (n = 16), and vehicle control group (n = 5). Data are expressed as means ± standard deviation. ∗p < 0.05 versus control groups. STZ, streptozotocin.

Download full-size image

Fig. 4. Trends of thermal withdrawal threshold in diabetic group (n = 99), STZ control group (n = 16), and vehicle control group (n = 5). Data are expressed as means ± standard deviation. STZ, streptozotocin.

TR, NAPA, and TR/NAPA all produced dose-dependent antinociceptive effects in tail-flick assay with ED₅₀ values (95% confidence interval) of 8.40 (4.5–37.22) mg/kg for TR, 142.45 (95.1–251.73) mg/kg for NAPA, and 2.8/24.26 (0–5.49/0–37.61) mg/kg for TR/NAPA. Isobologram showed that ED₅₀ value of TR/NAPA mixture fell below the lower confidential limit line of additivity, suggestive of a superadditive antinociceptive effect between TR and NAPA in tail-flick assay (Fig. 5).

Download full-size image

Fig. 5. Interaction of TR and NAPA in tail-flick assay. Line AB represents the additivity line. The dash lines represent the confidence interval of additivity line. 50% Maximum antinociceptive effect value of TR/NAPA (marked by the red star) mixture falls below the lower limit of the confidence interval of additivity line. TR, Tramadol; NAPA, Acetaminophen.

TR, NAPA, and TR/NAPA all produced dose-dependent antinociceptive effects in electronic von Frey assay with ED₅₀ values (95% confidence interval) of 8.96 (5.48–23.15) mg/kg for TR, 123.44 (58.19–288.81) mg/kg for NAPA, and 4.9/42.42 (0–9.37/0–81.18) mg/kg for TR/NAPA. Isobologram showed that ED₅₀ value of TR/NAPA mixture fell in the confidential limit of additivity line, suggestting an additive antinociceptive effect between TR and NAPA in electronic von Frey assay (Fig. 6).

Download full-size image

Fig. 6. Interaction of TR and NAPA in electronic von Frey assay. Line AB represents the additivity line. The dash lines represent the confidence interval of additivity line. 50% Maximum antinociceptive effect value of TR/NAPA falls in the confidence interval of the additivity line. TR, Tramadol; NAPA, Acetaminophen.

4. Discussion

The present study demonstrated that diabetic rats induced by a single injection of STZ could develop significant hyperalgesia. Both TR and NAPA showed dose-depending antinociceptive effects on PDN rats, whereas the interaction between TR and NAPA was additive in electronic von Frey test and superadditive in tail-flick test.

The STZ-induced diabetic rat model has been extensively used in research works exploring the pathogenesis and complications of the disease.⁸ However, the dosage and administration route of STZ are inconsistent in different studies. Although a large dose or intravenous route is believed to be associated with a higher induction rate as compared with a small dose or intraperitoneal route, it may produce severe diabetes which can impair the general health status of the rats.⁹ The general health status of rats, such as severe dehydration, electrolyte imbalance, and severe muscle weakness, may have an effect on the nociceptive threshold. This study demonstrated that a dose of 60 mg/kg STZ, given intraperitoneally was an appropriate choice to produce diabetic animal model without impairing their general health.

The present study showed that STZ-induced diabetic rats developed significant thermal and mechanical hyperalgesia, whereas STZ-injected nondiabetic rats (STZ control group), like the vehicle control rats, exhibited no alteration in thermal and mechanical nociception. The results suggested that the cause of hyperalgesia in model rats would be the DM disease per se rather than the inherent neurotoxicity of STZ, although some authors have suggested that STZ may induce a variety of pathophysiological symptoms leading to altered nociceptive responses in animal models.10, 11

The quality and degree of diabetic neuropathic pain vary a lot both in clinical patients¹²and animal models. Animal models of PDN performed by different laboratories showed different characteristics such as the amplitude and time course of hyperalgesia.13, 14 In our study, thermal and mechanical hyperalgesia were present on the fourth day after STZ injection, most prominently on the fourth week after STZ injection. A decrease of 38.3% in TFL and 31% in MWT was observed in the fourth week as compared with baseline values. These results are consistent with most previous studies.¹⁴

The alteration pattern of TWT in diabetic rats remains inconsistent among previous studies, and both decrease and increase trends in TWT have been observed.15, 16, 17 While in our study, although diabetics rats showed a significant decrease in TWT as compared with the basal value, rats in two control groups showed a similar decrease trend with no significant differences among the diabetic group and the control groups. The reason why TWT shows an alteration pattern different from TFL and MWT is still unknown. But TWT tested in hot-plate test may be affected by the animal’s body weight. The significant increase in body weight may result in the decrease in TWT in control rats, whereas the influence of body weight on TWT is minor in diabetic rats. The influence of body weight may diminish the difference in TWT between diabetic group and control groups. Because the measurement of TWT might be affected by other reasons, drug effects were not accessed by hot-plate test.

It is well established that diabetic neuropathic pain does not respond well to the routine analgesics, including opioids and NSAIDs. Although the exact mechanism of drug resistance to PDN is unknown, studies have shown that excitation of N-methyl-D-aspartate (NMDA) receptor,¹⁸ decrease in serotonin (5-HT) levels in brain nucleus,19, 20 and impaired effect of GABAnergic system may be underly to reduce efficacy and rapid drug tolerance of opioids. And analgesic efficacy of noradrenaline/5-HT reuptake inhibitors has been validated in STZ-induced PDN rat models.²¹ TR is a low-efficacy opioid with the mechanism of inhibiting 5-HT and NA reuptake. The unique analgesic mechanism makes TR a potent drug for the treatment of PDN. And this hypothesis has also been proved by a randomized controlled clinical trial.²² NAPA is another novel analgesic, which may possibly have several mechanisms, including the inhibition of NMDA receptor activation,²² 5-HT/NA reuptake,23, 24 and endocannabinoids uptake.²⁵ Pharmacological studies suggest that if the mechanism of action of either drug in a mixture is mediated through different receptors, the two drugs may have either an additive or superadditive interaction.²⁶ Therefore, NAPA is expected to enhance the analgesic effect of TR theoretically. This has been validated in the treatment of chronic pain associated with osteoarthritis,²⁷ postsurgical pain,²⁸ and oral surgery.²⁹ In the present study, we extended the application of TR/NAPA combination to the treatment of chronic PDN, and assessed the characteristics of drug interactions between the two drugs. Previous research has demonstrated that the nature of drug interactions is also dependent on the relative proportion of each drug in the mixture.³⁰ In agreement with this finding, another study revealed that the mixtures with the proportion of TR to NAPA lower than 1–5.7 produced synergistic effects, whereas mixtures with a dose ratio of 1: 8.667, 1:19, or 1:50 exhibit synergistic effects of statistical significance. Therefore, the TR/NAPA mixture with a dose ratio of 1: 8.667 was tested in the present study. And the results showed that the interaction of the two drugs in tail-flick assay was superadditive, whereas the interaction in electronic von Frey assay was additive. The different results of drug interaction between the two behavioral tests may have several reasons. First, measurement of nociceptive threshold in model rats can be affected by numerous factors, despite the correct method of measurement and the strict control of experimental environment. Therefore, the limitation of animal study may impair the accuracy of the results. Second, thermal and mechanical hyperalgesias in diabetic rats may have different mechanisms, therefore their response to drugs may be different.

In conclusion, we demonstrated that TR and NAPA could produce synergistic analgesic effect in PDN rats. This finding is consistent with the only clinical trial available on evaluation of efficacy of TR/NAPA in treating PDN patients,²² which indicates that a large-scale controlled trial of TR/NAPA in patients with PDN is warranted.

^†Declaration: Financial supports of this study were from the Department of Anesthesiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences.