Comparative Study between Preemptive and Postoperative Intra-Articular Injection of Levobupivacaine and Tramadol for Control of Postoperative Pain.

BACKGROUND: This study had been formulated to evaluate and compare the analgesic effect of preemptive (PE), postoperative (PO), and combined knee intra-articular injection (IAI) of levobupivacaine and tramadol after knee arthroscopy.
MATERIALS AND METHODS: A total of 220 patients assigned for therapeutic knee arthroscopy were divided into four equal groups. Patients in Group C received IAI of 20 mL (0.5%) levobupivacaine preoperative, meanwhile patients in Group PE received IAI of 18 mL (0.5%) levobupivacaine with 100 mg tramadol (2 mL). Patients in Group PO received IAI of 18 mL (0.5%) levobupivacaine with 100 mg tramadol (2 mL) postoperatively, whereas patients in Group PE/PO received IAI of 19 mL (0.25%) levobupivacaine with 50 mg tramadol (1 mL) preoperatively and postoperatively. Numeric rating scale (NRS) had been used to assess pain sensation. Duration till the first request of rescue analgesia and number of requests were recorded.
RESULTS: NRS scores were significantly higher in Group C compared to other groups and in Group PE in comparison to PO and PE/PO groups. Frequency of rescue analgesia requests was significantly higher in Group C with significantly higher mean times of requests, while was significantly lower in Group PE/PO with significantly lower mean times of requests when compared to groups PE and PO.
CONCLUSIONS: PE levobupivacaine and tramadol IAI provided satisfactory level of PO analgesia after therapeutic arthroscopy. However, combined PE and PO levobupivacaine and tramadol IAI of half dose provided PO analgesia superior to that provided by either PE or PO full-dose IAI.

INTRODUCTION

Arthroscopic knee surgery is very widely common orthopedic procedures which are performed in the outpatient setting. Although knee arthroscopy causes fewer traumas than open surgery, considerable PO pain can hinder patient's ability to participate in early rehabilitation and affect the patient's activity level and satisfaction. Patients are usually discharged shortly after surgery and must be provided with analgesia, which is both safe and effective.[1]

Intravenous (IV) opioid application is the treatment of choice in the immediate PO course. However, the use of opioid medication bears a risk of adverse effects such as nausea, vomiting, hypotonia, depression of respiratory center and excessive sedation. That is why, search for other methods able to affect nociceptive processes, for example, by peripheral route, is invariably in progress, especially for procedures of the lower extremity. Such assumption paved the way for the use of regional blocks,[2] and even local IAI of varied medications for PO pain management.[3]

The existence of opioid receptors in tissues outside the central nervous system constitutes an opportunity to administrate opioids peripherally to lessen as well acute (e.g., PO) as chronic pain with no risk of centrally mediated unacceptable effects. Single intra-articular administration of local anesthetic in combination with opioids has been used to provide better analgesia after arthroscopic knee surgery and to reduce consumption and possible side effects of oral and IV anesthetics or analgesics.[4]

Tramadol is an aminocyclohexanol derivative with μ-agonist activity and inhibits noradrenaline and 5-hydroxytryptamine neuronal uptakes thus prolonging the duration of their action. The combination of opioid and nonopioid mechanisms is believed to result in synergistic potentiation of tramadol analgesia.[5]

Local tramadol had been proved to be an effective modality for PO analgesia as documented by many authors such as Akkaya et al. who used peritonsillar tramadol infiltration,[2] Senel et al. who added tramadol to brachial plexus block, Ceccheti et al. who used submucosal tramadol injection after oral surgery, Afolayan et al. who used intrathecal tramadol for the management of visceral pain during subarachnoid block for appendectomy, and Alemanno et al. who found tramadol is effective as perineural adjuvant during arthroscopic rotator cuff repair under middle interscalene block.[6]

This study aimed to evaluate the efficacy of combined PE and PO knee IAI of levobupivacaine and tramadol on postarthroscopy pain versus PE or PO knee IAI only.

MATERIALS AND METHODS

The current prospective comparative study was conducted at Ain Shams University hospital from December 2016 to May 2018. The study protocol was approved by the Local Ethical Committee after obtaining written fully-informed patient's consents. The study is registered at clinical trials. Gov ID: NCT03785041.

The study included 220 patients fulfilling the inclusion criteria including patient physical status American Society of Anesthesiologists (ASA) classes I and II, aged between 18 and 60 and free of general diseases prohibiting the use of local anesthesia. Patients had allergy for local anesthetics or to tramadol, refusing participation in the study, with body mass index (BMI) >35 kg/m2 or had psychological or neurological disorders inducing difficulty for evaluation of pain scoring were excluded from the study.

Patients were randomly, using sealed envelopes chosen by the patient him/herself, divided into four equal study groups. Each group included 55 patients. Group C included patients assigned to receive IAI of 20 mL of 0.5% levobupivacaine (chirocaine 0.5%, 5 mg/ml, AbbVie Ltd., UK) alone preoperatively as PE analgesia (30 min before the procedure). Group PE included patients assigned to receive IAI of 18 mL of 0.5% levobupivacaine plus 100 mg tramadol (2 mL) as PE analgesia. Group PO included patients assigned to receive IAI of 18 mL of 0.5% levobupivacaine and 100 mg tramadol (2 mL) at the end of surgery as PO analgesia (10 min before the release of tourniquet), and Group PE/PO included patients assigned to receive IAI of 19 mL of 0.25% levobupivacaine (chirocaine 0.25%, 2.5 mg/ml, AbbVie Ltd., UK) in addition to 50 mg tramadol (1 mL) before the start of the surgery as PE analgesia and another 19 mL of 0.25% levobupivacaine plus 50 tramadol (1 mL) at the end of the surgery as PO analgesia [Figure 1].

Figure 1

Diagram displaying the recruitment of the patients into the study

Patients were premedicated 30 min before arrival to theater using midazolam tablet 7.5 mg and local application of eutectic mixture of local anesthetics cream at the sites that will be used as injection sites. On arrival to the operating room, standard monitoring with noninvasive monitoring of blood pressure, heart rate (HR), and peripheral oxygen saturation was applied for all patients. Lactated Ringer's solution was started and patients were given prophylactic IV broad-spectrum antibiotic before the start of the procedure. The patients in Group C, PE, and PE/PO were given IAI according to the medication designed for each group. The patients in Group C were given IAI of 20 mL 0.5% levobupivacaine. Furthermore, the patients in Group PE were given IAI of 18 mL 0.5% levobupivacaine with 100 mg tramadol (2 mL), whereas the patients in PE/PO group were given IAI of 19 mL 0.25% levobupivacaine in addition to 50 mg tramadol (1 mL).

All patients were operated upon under general anesthesia. After oxygenating the patient with 100 % oxygen for 3 min, anesthesia induced by injecting the patient with fentanyl 1 ug.kg-1 followed by propofol 2 mg.kg-1. i-gel supraglottic airway (i-gel intersurgical) was inserted for all patients. Anesthesia maintained by sevoflurane inhalation with minimal alveolar concentration. At the end of the surgery and before the release of tourniquet, the patients in Group PO were given IAI with 18 mL 0.5% levobupivacaine plus 100 mg tramadol (2 mL). The patients in PE/PO Group were given IAI with 19 mL 0.25% levobupivacaine plus 50 mg tramadol (1 mL). At least 10 min after IAI the tourniquet released, the patient started to be awakened, i-gel removed and the patient shifted to postanesthesia care unit. The medications for injection were prepared, labeled, and sealed with anesthesia registrar who was not oriented by the study. All knees IAI were given by the surgeon who was unaware of the study.

Pain sensation was assessed using a 11-point numeric rating scale (NRS) with numbers from 0 to 10 where 0 indicates no pain and 10 indicates worst pain imaginable. NRS was chosen for being more practical than the graphic visual analog scale, easier to understand for most people, and does not need clear vision, dexterity, paper and pen.[67]

PO pain sensation was assessed every 30 min for 2 h and hourly till 12 h postoperatively. Duration of PO analgesia was defined as duration since baseline PO pain evaluation till the request of rescue analgesia. Efficacy of analgesia was defined as the number of requests of rescue analgesia till discharge. Rescue analgesia was provided as IV lornoxicam 8 mg to guard against sedation or drowsiness as surgery was performed as day surgery. Lornoxicam 8 mg IV was administered for only two times as rescue analgesia. If the patient needed rescue analgesia for the third time, so we would provide 50 mg intramuscular pethidine.

During the peri and PO period, the standard monitoring including HR, systolic, diastolic and mean arterial pressure were recorded after completion of injection and every 5 min till 15 min and then every 15 min till the end of the surgery.

Patient's satisfaction rates concerning the efficacy of PO analgesia were assessed using the Iowa Satisfaction with Anesthesia Scale (ISAS) for monitored anesthesia care (MAC).[8] ISAS was documented as a valid tool for the Arabic speaking patients by Baroudi et al.[9] Patients respond to 11 statements by placing a mark along a six-choice vertical response column, each had numerical correspondence as follows: disagree very much (=1), disagree moderately (=2), disagree slightly (=3), agree slightly (=4), agree moderately (=5), and agree very much (=6). The scores were reversed in negative questions. The mean of their responses to each of the 11 statements gives a single number which is a quantitative measure of a patient's satisfaction with their MAC in a range of 11 meaning disagree very much to 66 means agree very much.

Statistical analysis

The sample power was calculated according to Kraemer and Theimann,[10] using their proposed figure showed that the sample size for 60% power would require an N of 31/group and 80% power would require an N of 51/group. This hypothesis was documented by Murphy and Myors.[11] Considering therapeutic arthroscopy is going to be frequent procedure and performing it under local anesthesia, allowed it to be managed as one-day surgery, so from a standard nomogram, a sample size of >51 patient per group was determined to be sufficient to give the trial >80% power to detect a difference at the 5% significance level. Obtained data were presented as mean ± standard deviation, ranges, numbers, and ratios. Results were analyzed using paired t-test for within groups variability, Wilcoxon ranked test for non-parametric related data (Z-test) and Chi-square test (χ2 test) for categorical variables between groups. Statistical analysis was conducted using the SPSS 20 (IBM, Armonk, NY, USA) for Windows statistical package. P < 0.05 was considered as statistically significant.

RESULTS

This study included 220 patients assigned for therapeutic knee arthroscopy for variant indications; 107 patients (48.6%) had arthroscopy for traumatic knee indications and the remaining number for inflammatory conditions. All data including age, sex, body weight and BMI are represented in Table 1. One-hundred and fifty-two patients (69%) were ASA Grade I and 88 patients (31%) were ASA Grade II. Operative procedures varied according to the indication and mean operative time were 50.8 ± 9.5; range: 40–80 min. There was nonsignificant (P > 0.05) difference between studied groups as regard enrolment data, clinical and operative data [Table 1].

Table 1

Enrolment data of studied groups

	Group C	Group PE	Group PO	Group PE/PO	P
Age (years)	31.4±8.9	30.9±8.1	30±7.1	28.6±8.8	0.304
Gender, n (%)
Male	34 (61.8)	36 (65.5)	31 (56.4)	33 (60)	0.803
Females	21 (38.2)	19 (34.5)	24 (43.6)	22 (40)
Body weight (kg)	87.9±9.1	86.1±8.7	89±8.5	85.5±10.6	0.1729
Body height (cm)	170.6±3.9	170.2±3.6	170.3±4	170.1±3.6	0.9101
BMI (kg/m2)	30.2±3.4	29.7±3	30.7±3.7	29.6±4.3	0.3624
ASA physical status, n (%)
I	36 (65.4)	38 (69.1)	41 (74.5)	37 (67.2)	0.375
II	19 (34.6)	17 (31.8)	14 (25.4)	18 (32.7)	0.432
Diagnosis, n (%)
Osteoarthrosis	13 (23.6)	15 (27.3)	13 (23.6)	14 (25.5)	0.585
Meniscal injury	10 (18.2)	13 (23.6)	14 (25.5)	9 (16.4)	0.293
Hypertrophic synovitis	12 (21.8)	10 (18.2)	11 (20)	11 (20)	0.865
Cruciate ligament injury	9 (16.4)	7 (12.7)	8 (14.5)	9 (16.4)	0.781
Chondral injury	7 (12.7)	8 (14.6)	6 (10.9)	7 (12.7)	0.642
Plica synovialis mediopatellaris	4 (7.3)	2 (3.6)	3 (5.5)	5 (9.1)	0.317
Procedure, n (%)
Partial menisectomy	23 (41.8)	18 (32.7)	19 (34.5)	21 (38.2)	0.680
Partial synovectomy	12 (21.8)	17 (30.9)	16 (29.1)	15 (27.3)	0.412
Chondroplasty	11 (20)	9 (16.4)	9 (16.4)	10 (18.2)	0.794
Resection of the plica	6 (10.9)	7 (12.7)	8 (14.5)	7 (12.7)	0.801
Removal of free body	3 (5.5)	4 (7.3)	3 (5.5)	2 (3.6)	0.514
Duration of surgery (min)	58.1±14.1	58.4±18.8	62.4±21.2	62.7±18.5	0.387

Data are presented as mean±SD and numbers; percentages are in parenthesis. SD=Standard deviation, PE=Preemptive, PO=Postoperative, BMI=Body mass index, ASA=American Society of Anesthesiologist

Mean duration of PO analgesia was significantly (P < 0.05) shorter in Group C compared to other groups; but was significantly (P < 0.05) longer in Group PE/PO compared to that of Groups PE and PO and nonsignificantly (P > 0.05) longer in Group PO compared to that of Group PE [Figure 2].

Figure 2

Mean (±standard deviation) duration of postoperative analgesia

A total PO pain NRS scores was significantly (P < 0.05) higher in Group C compared to other groups with significantly higher scores of patients of Group PE compared to PO and PE/PO groups. However, patients of the PO group showed nonsignificantly (P > 0.05) higher pain scores compared to patients of PE/PO group [Figure 3].

Figure 3

Mean (±standard deviation) postoperative numeric rating scale score of studied groups

The frequency of patients requested rescue analgesia for more than once was significantly higher in Group C with significantly higher mean times of requests compared to other groups. While the frequency of patients requested rescue analgesia for more than once was significantly lower in Group PE/PO with significantly lower mean times of requests compared to groups PE and PO that showed nonsignificant difference in favor of PO group [Table 2 and Figure 4].

Table 2

Data of postoperative analgesia of studied groups

	Group C	Group PE	Group PO	Group PE/PO	P
Duration of PO analgesia (min)	3.5±1.3	5.3±1.4*,‡	5.6±2.1*,‡	6.4±1.5*	<0.001
PO pain NRS score	2±0.4	1.6±0.4*,†,‡	1.4±0.4*	1.3±0.5*	<0.001
Frequency of requests of rescue analgesia (%)
Once	3 (5.5)	21 (38.2)	29 (52.7)	37 (67.3)
2-times	46 (83.6)	32 (58.2)	23 (41.8)	17 (30.9)
3-times	6 (10.9)	2 (3.4)	3 (5.5)	1 (1.8)
Mean times of requests	2.05±0.4	1.65±0.55*,‡	1.53±0.6*,‡	1.35±0.5*	<0.001

Data are presented as mean±SD. *Significant difference versus Group C, †Significant difference versus Group PO, ‡Significant difference versus Group PE/PO. SD=Standard deviation, PE=Preemptive, PO=Postoperative, NRS=Numeric rating scale

Figure 4

Mean (±standard deviation) times of requests of rescue analgesia by patients of studied groups

All patients were satisfied by the type of IA analgesia provided; however, the frequency of patients among higher satisfaction scores was significantly (P < 0.05) lower in patients of Group C compared to other groups that showed nonsignificant difference, despite being in favor of Groups PO and PE/PO. Calculated mean satisfaction score was significantly lower in group C compared to other groups and also in group PE compared to group PO, while it was with higher scores in group PO compared to group PE/PO and in group PE/PO compared to group PE but with non-significance (P > 0.05) [Table 3 and Figure 5].

Table 3

Patients’ satisfaction by postoperative analgesia of studied groups

Frequency of patients among satisfaction scores	Group C (%)	Group PE (%)	Group PO (%)	Group PE/PO (%)	P
11-	0	0	0	0
22-	8 (14.5)	7 (12.7)	3 (5.5)	5 (9.1)
33-	24 (43.6)	14 (25.5)	9 (16.4)	11 (20)
44-	15 (27.3)	21 (38.2)	25 (45.5)	23 (41.8)
55-66	8 (14.5)	13 (23.6)	18 (32.6)	15 (29.1)
Mean score	42.6±10.4	46.8±9.8*,†	50.8±9.2*	49.1±9.4*	0.0314

Data are presented as mean±SD. *Significant difference versus Group C, †Significant difference versus Group PO. SD=Standard deviation, PE=Preemptive, PO=Postoperative

Figure 5

Mean (± standard deviation) patients’ satisfaction score by provided analgesia

DISCUSSION

The obtained results indicated the efficacy of IAI of levobupivacaine and tramadol as an analgesic modality for patients undergoing therapeutic arthroscopy, irrespective of timing of injection, manifested by significant difference in comparison to control patients as regard total NRS pain scores, duration of analgesia, frequency of consumption of rescue analgesia, and patient's satisfaction scoring. These findings supported what previously reported in the literature concerning the efficacy of local administration of tramadol in addition to local anesthetics and assured the applicability of IAI of tramadol for provision of peri-operative analgesia that lasts for ≥ 6 PO hours.

Concerning drug choice to be used for IAI, the use of tramadol was and still a point of controversy. Beyzadeoglu et al. found IA tramadol plus periarticular bupivacaine provides a better pain relief and less analgesic requirement, following arthroscopic outpatient partial meniscectomy compared to IA bupivacaine.[12] Furthermore, Zeidan et al. found IAI of tramadol 100 mg with bupivacaine 0.25% provides a pronounced prolongation of analgesia compared with either drug alone in patients undergoing day-care arthroscopic knee surgery.[13]

On contrary, Ayoglu et al. documented that administration of IA tramadol-ketamine combination was found to be more effective in decreasing PO daily analgesic consumption compared to tramadol alone.[14]

Thereafter, Mitra et al. found pain scores after IAI of bupivacaine with either fentanyl or tramadol was better than IA bupivacaine alone, and despite fentanyl was better, both combinations were comparable to each other with regard to supplementary analgesic requirement.[15] Arti and Mehdinasab compared IAI of morphine, pethidine, methadone, and tramadol on PO pain after arthroscopic anterior cruciate ligament reconstruction and found the lowest significant pain intensity was with morphine, while morphine and methadone groups needed the lowest dosage of analgesics.[16]

On the other hand, Jazayeri et al. found no statistically significant difference in PO pain scoring between IAI of morphine and tramadol except for 1-h PO scores in which tramadol-treated group experienced less pain and documented a PO analgesic effect of IA morphine and tramadol following minor arthroscopic knee surgeries with a maximum effect 6-h postinjection.[17]

Multiple studies tried to explore the mechanism of action of locally administered tramadol. Early studies by Bianchi et al. reported significant decrease of synovial fluid concentrations of substance P, and nonsignificant decrease in concentration of IL-6 and proposed that the activity of tramadol may involve the modulation of inflammatory nociceptive mediators.[18] Altunkaya et al. assumed a local anesthetic effect of tramadol through comparison between subcutaneous injection of tramadol versus lidocaine during minor surgical procedures and reported nonsignificant difference as regard operative pain sensation and patient's satisfaction, so proposed that tramadol can be used as an alternative drug to lidocaine for minor surgeries.[19] Alagöl et al. reported that tramadol handovers analgesia with a peripheral mechanism when administered intraarticularly.[20]

Thereafter, Kimura et al. indicated that tramadol inhibits PO hypersensitivity by increasing noradrenaline (NA) and serotonin (5-HT) transmission levels in the spinal cord in addition to activating opioid receptors and concluded that tramadol might be more effective during early PO period when spinal NA and 5-HT levels are increased.[21]

Sousa and Ashmawi experimentally concluded that tramadol conferred an early local analgesic action and lessening mechanical hyperalgesia produced by plantar incision, but this analgesic effect was not interceded by peripheral opioid receptors.[22] Singh et al. appraised the modifications in PO serum IL-6 levels succeeding surgical removal of third molars under local anesthesia after administration of diclofenac, ketorolac, and tramadol and mentioned that elimination of IL-6 values arise in tramadol group intimately following the diclofenac group and though ketorolac suppresses IL-6 levels similar to diclofenac initially but after 7 days tramadol and ketorolac suppressed IL-6 expression in similar manner which is less compared to diclofenac group.[23]

Concerning timing of injection, PE tramadol IAI in Group PE showed nonsignificant difference of total NRS pain scores compared to PO injection. However, the significantly shorter duration of analgesia and frequency of requests of rescue analgesia could be attributed to the consumption of analgesic effect throughout operative time that extended for about 50 min, thus the duration of PE analgesia could be considered longer than or at least similar to duration provided by PO analgesia. These findings indicated the efficacy of PE tramadol analgesia in minimizing PO or postprocedural pain irrespective of clinical situation and were in line with Tuncer et al. who found analgesic duration, total analgesic consumption, and PO pain scores recorded at rest and with movement were significantly lower and patient satisfaction was significantly higher in patients received PE or PO tramadol compared to controls and with PE compared to PO tramadol.[24] Koputan et al. who indicated that PE tramadol infiltration technique is an efficient, practical, and safe alternative to levobupivacaine in septorhinoplasty or endoscopic sinus surgery,[25] da Costa Araújo et al. who documented that peak pain occurred 5 h after bilateral lower third molar surgeries in patients received PE tramadol infiltration,[26] and Jabalameli et al. who documented that subcutaneous PE infiltration of tramadol before surgical incision reduces PO opioid consumption.[27] Furthermore, Jensen et al. concluded that tramadol administered 30 min before capsaicin 8% patch application should be treated as a different pretreatment option in patients receiving capsaicin patch treatment.[28]

One of the interesting findings of the current study is the outcome of patients of Group PE/PO, who received combined PE and PO levobupivacaine and tramadol IAI of half the dose and showed significantly longer duration of analgesia and nonsignificantly lower total NRS scores and frequency of requests of rescue analgesia compared to those received PO analgesia and these parameters were superior to that provided by total dose PE analgesia.

Limitations

It would be more valuable if we could study different concentrations of tramadol and levobupivacaine to set the optimum dose range to control pain, but this needs more wide scale of patients and studies. In this study, it would be more feasible if we could increase another control group for levobupivacaine to be injected intraarticular PO, but we found that it would be more confusing but this may be done in a separate study. It was better if all IAI to be given by the same surgeon, but we failed to fulfill this condition due to large number of cases included in the study.

CONCLUSION

It could be concluded that PE levobupivacaine and tramadol IAI provided satisfactory level of PO analgesia after therapeutic arthroscopy comparable to PO administration. However, combined PE and PO levobupivacaine and tramadol IAI of half-dose provided PO analgesia superior to that provided by either PE or PO full-dose IAI. Wider scale studies are mandatory to establish outcome of combined-reduced dose of levobupivacaine and tramadol IAI.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.