Retrospective Analysis of Pain Relief in Total Knee Replacement Surgeries.

BACKGROUND AND AIMS: The aim of the study is to measure the postoperative adequacy of pain relief and functional recovery after unilateral primary total knee arthroplasty or total knee replacement (TKR) with a multimodal approach. SETTINGS AND
DESIGN: This was a retrospective observational study done in a tertiary care center.
MATERIALS AND METHODS: Eighty patients aged 18-65 years (yrs) with ASA Physical Status Classes I, II, and III operated for unilateral primary TKR surgery under suitable Anaesthesia and was administered ultrasound-guided adductor canal block + periarticular infiltration (PI) from January 2018 to January 2019 were included. Thereafter, the patients visual analog scale (VAS) scores at rest, 45° knee flexion, and mobilization as well as additional analgesia given were noted after going through the records at following time points: 12 hourly for 24 h and thereafter on postoperative day 1 (POD1) and POD2. Level of block, adverse events, and functional recovery (time up and go [TUG] test, 10 s walk test) on POD1 and POD2 were also noted. STATISTICAL ANALYSIS AND
RESULTS: The statistical software namely SPSS 18.0 were used for the analysis of the data. The mean VAS score at 12 h was 4.33 ± 1.3 which worsened at 24 h with steady improvement over the next 2 days. Similarly, the case with a mean VAS score at 45° flexion and on mobilization showed a similar trend. At 12 h postadductor block, besides intravenous (i.v.) paracetamol, 32.6% of patients were given tramadol 50 mg i.v. whereas one patient was given a buprenorphine patch in addition to tramadol. The number of patients requiring additional tramadol and buprenorphine patch steadily increased over the next 2 days. The average time taken for the TUG test at 24 h was 30.98 ± 4.77 s, and the average time taken for the 10 s walk test at 24 h was 6.16 ± 1.10 steps with improvement in performance over the next 2 days.
CONCLUSION: In our study, our multimodal analgesia model did not provide satisfactory analgesia though mobilization was not hampered. Copyright:

INTRODUCTION

Total knee replacement (TKR) is regarded as an effective treatment for end-stage knee osteoarthritis.[1] Total knee arthroplasty (TKA) or TKR involves extensive bone resection and soft tissue manipulation, and patients can experience severe pain during the early postoperative period. Appropriate pain management after TKA allows faster recovery, reduces the risk of postoperative complications, improves patient satisfaction, and decreases incidence of the development of chronic pain syndrome complex.[2] There is an increasing focus on improved pain control, while maintaining optimal patient safety without affecting postoperative patient's mobility and physiotherapy. Femoral nerve block (FNB) reduces the strength of the quadriceps muscle and increases the risk of falling after TKA which can lead to periprosthetic fracture and other complications.[34] Hence, focus is on an effective analgesic modality where the motor strength is preserved to enable faster rehabilitation. Ultrasound (USG)-guided adductor canal block (ACB) is one such technique which blocks the saphenous nerve which is purely sensory. However, a multimodal approach is necessary for optimum analgesia. In this study, we aimed to measure the adequacy of pain relief with a multimodal approach in the postoperative period after TKR surgeries (primary objective). The functional recovery and the need for additional analgesia were also studied (secondary objective).

MATERIALS AND METHODS

After Institutional Review Board and Ethics Committee Approval, 80 consenting patients of either sex, aged 18–65 years with American Society of Anesthesiologists (ASA) Physical Status Classes I, II, and III who were operated for unilateral primary TKR surgery in orthopedic operation theater in a tertiary care center and were administered ACB + PI from January 2018 to January 2019 were included in this retrospective observational study. Their preoperative assessment charts were studied for demographic data and presence of comorbidities. The type of anesthesia given, regional or general anesthesia, was noted. They were observed for pain relief and functional mobility. Additional analgesia in the form of ketorolac and pregabalin or any opioid administered was noted. The institutional protocol was to give spinal anesthesia for surgery with PI intraoperatively (20 mL 0.5% Bupivacaine, 80 mL saline 0.9%, 50 μg clonidine and 100 μg fentanyl) by a standard technique administered by senior operating surgeon. Ultrasound-guided ACB with 20 mL of 0.75% ropivacaine was given at the end of surgery by the senior anesthetist in the theater (considered as 0 h). This was administered by the standard technique that is at the middle third of the thigh, with drug injected perineurally to saphenous nerve lying superolateral to the femoral artery. Ultrasound imaging was done using Samsung Sonoace R7 (Manufacturer Samsung Medison Co, Republic of Korea) and linear frequency probe (7–13 Hz).

Postoperatively, all patients were given 1 g i.v. paracetamol 6 hourly and i.v. Tramadol 12 hourly. Thereafter, the patients were monitored 12 hourly for 24 h and thereafter on postoperative day (POD) 1 and day 2 for the following:

Visual analog scale (VAS) 0–10 at rest

VAS 0–10 on 45° knee flexion

Additional analgesia administered was noted after going through the records

Level of block (spinal/combined spinal epidural)

Adverse events: Local anesthetic systemic toxicity, quadriceps weakness, fall during mobilization, and others

Functional recovery on POD 1 and day 2 (performed after the spinal block recedes completely).

Time up and go test (TUG test): Used to assess patient's mobility. It requires their static and dynamic balance. It is the time taken by a person to rise from a chair, walk 3 m, turn around, walk back to the chair, and sit down. It is counted as seconds

10 s walk test measured as “number of steps.”

The above protocol was uniform for all patients.

Statistical analysis

The sample size of 80 has been calculated based on the available reference studies, within 95% confidence limit and 80% of power. Descriptive and inferential statistical analysis has been done in the present study. Results on continuous measurements are presented on mean ± standard deviation (SD) (minimum–maximum) and the results on categorical measurements are presented in number (%). Significance is measured at 5% level of significance.

Paired proportion test has been used to find the significance of proportion in paired data. Statistical software namely SPSS 18.0 (IBM company, Armonk, New York, USA), were used for the analysis of the data, and Microsoft word and Excel have been used to generate graphs, tables, etc.

RESULTS

This is a retrospective observational clinical study. The mean age was 55.64 ± 8.13 (mean ± SD) years. The maximum patients were in the age group of 51–65 years (68.8%). The proportion of females (66.3%) was much more compared to males.

Hypertension was the most common systemic disease affecting our study group (38.8%) followed by diabetes mellitus (10%) and rheumatoid arthritis (7.6%). However, 2.5% of patients were obese. Those having bronchial asthma, ischaemic heart disease, tuberculosis, and valvular heart disease were 1.3% each. Majority belonged to the ASA Physical Status Class II (65%) whereas 2.5% belonged to ASA Physical Status Class III.

Mean VAS score at 12 h postblock was 4.33 ± 1.3. Fifty-two percent of patients had VAS scores <5, 12 h postblock. VAS scores worsened at 24 h (22.5% with VAS below 5) and the mean VAS score was 5.14 ± 1.08. There was some improvement in mean VAS scores over the next 2 days probably due to decrease in the tissue edema.

Only 38.8% of patients had VAS scores <5 at 45° flexion at 12 h. The mean VAS score at 12 h on 45° flexion was 4.94 ± 1.43 and at 24 h was 6.04 ± 0.95.

Mean VAS scores on mobilization at 12 h were 4.96 ± 1.37 and at 24 h were 6.04 ± 1.04. 61.3% of patients had VAS scores above 5 at 12 h, whereas 93.8% of patients had VAS scores above 5.

The assessment made at various study points 0, 12 h, 24 h, POD1, and POD2 using paired proportion test suggested a significant (P < 0.001) increase in the number of patients with VAS score of >5 at rest over time. This indicated insufficient pain relief with our multimodal approach. Similar results were obtained with VAS scores at 45° flexion and on mobilization [Table 1 and Figure 1].

Table 1

Visual analog scale scores

Time since ACB	Number of patients with VAS
	<5, n (%)	>5, n (%)
VAS at rest
0 h	80 (100)	0
12 h	42 (52.5)	38 (47.5)
24 h	18 (22.5)	62 (77.5)
POD 1	20 (25)	60 (75)
POD 2	27 (33.8)	53 (66.3)
Percentage difference	−66.2	66.2
P	<0.001
VAS at 45 flexion
0 h	80 (100)	0
12 h	31 (38.8)	49 (61.3)
24 h	2 (2.5)	78 (97.5)
POD 1	7 (8.8)	73 (91.3)
POD 2	10 (12.5)	70 (87.5)
Percentage difference	−87.5	87.5
P	<0.001*
VAS on mobilization
0 h	80 (100)	0
12 h	31 (38.8)	49 (61.3)
24 h	5 (6.3)	75 (93.8)
POD 1	8 (10)	72 (90)
POD 2	10 (12.5)	70 (87.5)
Percentage difference	−87.5	87.5
P	<0.001*

*P is significant. VAS=Visual analog scale, POD=Postoperative day, ACB=Adductor canal block

Figure 1

Visual analog scale scores (mean ± standard deviation)

At 12 h postadductor block, besides i.v. paracetamol, 32.6% of patients were given tramadol 50 mg i.v. whereas one patient was given a buprenorphine patch in addition to the above. 66.3% of patients required no additional analgesia.

At 24 h, 47.6% of patients were given tramadol, whereas 7.5% needed buprenorphine patch in addition to tramadol. On POD1, 57.5% were given tramadol whereas 15% received a buprenorphine patch in addition. On POD2, 58.8% received tramadol whereas 17.5% received buprenorphine patch in addition [Table 2].

Table 2

Opioid analgesia

Time	Number of patients (n=80), n (%)
Zero hours
Nil	80 (100.0)
Yes	0
12 h
Nil	53 (66.3)
Tramadol	26 (32.6)
Buprenorphine patch	1 (1.3)
24 h
Nil	36 (45.0)
Buprenorphine patch	6 (7.5)
Tramadol	38 (47.6)
POD 1
Nil	22 (27.5)
Buprenorphine patch	12 (15.0)
Tramadol	46 (57.5)
POD 2
Nil	19 (23.8)
Tramadol	47 (58.8)
Buprenorphine patch	14 (17.5)

POD=Postoperative day

Time up and go test

At 12 h, only eight patients were mobilized, of them four required <30 s and the other four required 30–60 s. At 24 h, 21.3% required <30 s, and 78.8% required 30–60 s. On POD2, 46.3% required <30 s, and 53.8% required 30–60 s. On POD3, 58.8% required <30 s, and 41.3% required 30–60 s [Table 3 and Figure 2].

Table 3

Tug test (time in seconds)

Tug test	Minimum-maximum	Mean±SD
0 h	0.00-0.00	0.00±0.00
12 h	-	-
24 h	20.00-60.00	30.98±4.77
POD 1	15.00-40.00	29.08±4.19
POD 2	20.00-42.00	27.93±4.24

POD=Postoperative day, SD=Standard deviation

Figure 2

Time up and go test (time in seconds)

Ten-second walk test

At 12 h, the eight patients who were mobilized, they took 5–8 steps during the 10-s walk test. Over the next 2 days, all the patients were able to perform the above mobilization tests with an average of 5–8 steps [Table 4 and Figure 3].

Table 4

10-s walk test (mean number of steps)

10-s walk test	Minimum-maximum	Mean±SD
0 h	0.00-0.00	0.00±0.00
12 h	-	-
24 h	3.00-9.00	6.16±1.10
POD 1	4.00-9.00	6.50±1.08
POD 2	4.00-9.00	6.89±1.10

POD=Postoperative day, SD=Standard deviation

Figure 3

Ten-second walk test

Thus, there was significant improvement in performance of above tests over time.

None of the patients had nausea, dizziness, or respiratory depression or any other adverse events.

DISCUSSION

TKA is a successful intervention for patients with painful degenerative diseases affecting the knee joint. However, this procedure is known to cause moderate-to-severe pain and is hence regarded as one of the most painful orthopedic procedures.[5] The pain management after TKA has always been a key focus in the clinical treatment of patients undergoing this procedure. Postoperative pain leads to decreased ability to mobilize the knee, prolonged hospitalization, and increased complications. Despite comprehensive multimodal analgesic regimens, this problem has not been successfully addressed.[6] The traditional analgesic techniques for TKR are intravenous patient-controlled analgesia (i.v. PCA) with opioids, epidural analgesia (EA), and FNB. PCA cause nausea, vomiting, and pruritus. EA can be associated with hypotension, urinary retention, and pruritus. FNB was very popular but was associated with decreased quadriceps strength, hampering the mobilization and subjecting the patient to risk of fall.[7]. Fall after TKR can have devastating consequences including periprosthetic fractures or soft tissue damage. Preventing this is of utmost importance. This has increased the popularity of ACB as it selectively blocks saphenous nerve which is purely sensory.[4]

The adductor canal, also known as Hunter's canal or the sub-sartorial canal, is an aponeurotic intermuscular tunnel in the middle third of the thigh. The canal is triangular in cross-section. It is bounded by three muscles: quadriceps anterolaterally (specifically vastus medialis), sartorius medially, and adductor magnus posteriorly. Within this canal is the femoral artery, femoral vein, the posterior branch of the obturator nerve, and fascicular branches of the femoral nerve; specifically the saphenous nerve and nerve to vastus medialis.[8910] Both the saphenous nerve and nerve to vastus medialis contribute to the innervation of the anteromedial knee joint and are therefore important targets of ACB. Both nerves exit in the distal third of the adductor canal. Hence, the mid-portion of the adductor canal is an optimal site of local anesthetic administration to block both target nerves, while minimizing the possibility of proximal spread to the femoral triangle[11] [Figure 4].

Figure 4

Adductor canal sonoanatomy. *A = Artery*

However, the block is distal to most of the efferent branches to the quadriceps muscle and therefore largely preserves the strength of this muscle.[6] USG-guided ACB is now a part of every multimodal postoperative analgesia regimen.[5]

In our institute, TKR is done by medial parapatellar arthrotomy by senior surgeons. Along with the ACB, PI and 1 g i.v. paracetamol are our routine protocol. We reviewed all the records and summarized the analgesic efficacy of the above regimen, its effect on mobilization, and quadriceps strength.

We preferred to use 20 mL volume for the ACB in accordance tothe dose finding study by Jaeger et al.[12] Part of the knee joint is also supplied by branches of the sciatic nerve besides the branches of femoral and obturator.[8910] Administering a sciatic nerve block for the posterior part of the knee may cause foot drop impairing mobilization in the postoperative period. Furthermore, TKA-related injury has an incidence of 2.4% with many surgeons wanting to check its function in the postoperative period. Hence, it is not recommended.[13]

PI may be able to take care of pain in the posterior part of the knee joint without causing motor block. The technique used for PI was as per Kerr and Kohan.[14]

The mean age in our study was 55.64 ± 8.13 years (mean ± SD), whereas in the retrospective cohort study by Thacher et al., the average age was 69 years.[4]

The proportion of females was more, contributing 66.3% of the study group similar to finding of Thacher et al.[4]

The mean VAS scores (at rest) at 12 h and 24 h post blockwere 4.33 ± 1.32 (mean ± SD) and 5.14 ± 1.08 respectively. Whereas Hanson et al. had numerical rating scale (NRS) scores of 2 and 4 at 18 and 24 h, respectively.[15] The VAS and NRS scores are comparable and reproducible.[16]

Jenstrup et al., Jaeger et al., and Grevstad et al. evaluated the effect of ACB on pain during an active flexion of knee and at rest in patients after TKA, compared with saline injection, and found a significant reduction in pain scores during active flexion of knee.[171819] Yuan et al. in a meta-analysis found analgesia of ACB comparable to FNB.[20] Andersen et al. suggested that VAS pain scores during movement and rest on the day of surgery were significantly lower in the saphenous nerve block group but no improvement beyond day 1.[21] In a meta-analysis by Ma et al. wherein one group was given ACB +periarticular infiltration and the other was given only periarticular infiltration, found no difference in the VAS scores in the two groups.[7]

The mobilization attempted on the evening of surgery was bed-side standing. Only a few surgeons attempted mobilization with walker on the same evening. Average time of mobilization after block was 17 h. In spite of higher VAS scores in our patients (6.04 ± 1.04), they could perform mobilization. The VAS scores improved thereafter. Thacher et al. had mean VAS scores of 3.3 after physiotherapy.[4] Andersen et al. had a mean NRS score of 3 in the ACB + PI group on activity.[21]

Thatcher et al. compared the effect of FNB versus ACB and found opioid consumption to be significantly more in the ACB group on day 2. This was at par with our findings.[4] In the studies by Jenstrup et al. and Hanson et al., morphine consumption within first 24 h was significantly reduced in the ACB group compared that in the placebo group.[1517] Further, many of the above studies received various opioids by PCA i.v.[18] However, our patients received tramadol which was a mild analgesic. Most patients received the buprenorphine patch on day 2. This transdermal patch contains 10 mg of buprenorphine in a 12.5 cm2 area. It releases a trivial 10 μg of buprenorphine per hour over a period of 7 days. None of the patients had any adverse events following opioids.

ACB has a motor sparing action and largely preserves quadriceps strength, thus facilitating mobilization and preventing patient falls. Wang et al. and Thacher et al. found decreased fall rate and knee buckling after ACB compared to FNB.[14]

Wang et al. found better range of motion in ACB versus FNB but no difference in time to straight leg raising test.[1] Further, the ACB group could perform TUG test faster and more patients could do it in the first 24 h but no difference in the next 2 days. In the meta-analysis by Elazab et al. (ACB vs. sham catheters), they found no difference in the two groups with respect to the TUG test.[22] Jaeger et al. (continuous FNB vs. ACB) found a decrease in QMVIC (maximum voluntary isometric contraction of quadriceps) by 8% compared to 48% in FNB. Further, the performance of TUG test and 10-m walk test was decreased in the FNB group compared to the ACB group.[2324] The average time required by the ACB group to perform the TUG test in the above study was 37s at 24 h whereas in our study it was 30.98 ± 4.77 s.[23] Kwofie et al. and Hanson et al. also had similar findings.[1525] Ma et al. compared effect of ACB+PI versus only PI and found that the ACB+PI group could walk longer distances on POD1.[7]

Jaeger et al. and Hanson et al. found decrease in QMVIC even in the ACB group compared to baseline.[1523] No adverse events were noted in our records.

To summarize, pain relief was insufficient with our multimodal analgesic model. This did not affect the functional mobility which improved over time.

Drawbacks of our study were it was a retrospective analysis. In our setting, putting adductor canal catheters was not always possible considering the time and cost. The use of i.v. PCA pumps with opioids and adductor canal catheters with infusion pumps requiring additional monitoring was not feasible.

CONCLUSION

In our study, ACB with PI with i.v. paracetamol and tramadol did not provide satisfactory analgesia though mobilization was not hampered with the preservation of quadriceps strength. No adverse events occurred. ACB catheters would have been a preferred option but was not possible in our setting.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.