Reduction in opioid use with perioperative non-pharmacologic analgesia in total knee arthroplasty and ACL reconstruction: a systematic review

Introduction: The world’s opioid epidemic has gotten increasingly severe over the last several decades and projects to continue worsening. Orthopedic surgery is the largest contributor to this epidemic, accounting for 8.8% of postoperative opioid dependence cases. Total knee arthroplasty (TKA) and anterior cruciate ligament (ACL) reconstruction are commonly performed orthopedic operations heavily reliant on opioids as the primary analgesic in the peri- and immediate postoperative period. These downfalls highlight the pressing need for an alternate, non-pharmacologic analgesic to reduce postoperative opioid use in orthopedic patients. The presented systematic review aimed to analyze and compare the most promising non-pharmacologic analgesic interventions in the available literature to guide future research in such a novel field. Methods: A systematic search of PubMed, MEDLINE, Embase, Cochrane, and Web of Science was performed for studies published before July 2020 based on the PRISMA (preferred reporting items for systematic reviews and meta-analyses) guidelines, and the obtained manuscripts were evaluated for inclusion or exclusion against strict, pre-determined criteria. Risk-of-bias and GRADE (grades of recommendation, assessment, development, and evaluation) assessments were then performed on all included studies. Results: Six studies were deemed fit for inclusion, investigating three non-pharmacologic analgesics: percutaneous peripheral nerve stimulation, cryoneurolysis, and auricular acupressure. All three successfully reduced postoperative opioid use while simultaneously maintaining the safety and efficacy of the procedure. Discussion: The results indicate that all three presented non-pharmacologic analgesic interventions are viable and warrant future research. That said, because of its slight advantages in postoperative pain control and operational outcomes, cryoneurolysis seems to be the most promising. Further research and eventual clinical implementation of these analgesics is not only warranted but should be a priority because of their vast potential to reduce orthopedics surgeries’ contribution to the opioid epidemic.


Introduction
Total knee arthroplasty (TKA) and anterior cruciate ligament (ACL) reconstruction surgeries are two of the most performed orthopedic operations in the world [1][2][3][4]. TKA cases, revisions, and revisions per surgeon have all significantly increased in the last 10 years [2,5]; ACL reconstruction has similarly become more prevalent over the last 30 years, particularly in the pediatric population, corresponding with the United States' increase in youth sports injuries [1,3,6].
While both procedures are considered highly efficacious, with TKA having shown a 10-to 15-year implant survivorship rate of over 90% [2], and up to 92% of ACL reconstruction patients receiving "A" or "B" outcome scores in a 2018 systematic review by Sarraj et al. [6], they are both operations consistently associated with moderate to severe postoperative pain [1,4]. Exposed by this are the sub-par anesthetic and analgesic perioperative pain management interventions used in the field of orthopedic surgery.
According to Li et al. 2019 review, a multimodal approach is currently considered best practice for TKA pain management [7,8]. This approach subjects the patient to a wide variety of anesthetic and analgesic medications, including preemptive analgesia, neuraxial anesthesia, peripheral nerve blocks, patient controlled-analgesia, local infiltration analgesia, gabapentanoids, periarticular injections, non-steroidal anti-inflammatory drugs (NSAIDs), and opioids [7,8]. A similar multimodal approach is currently used peri-and postoperatively for ACL reconstructions, highlighted by intra-articular injections, peripheral nerve blocks, regional anesthesia, oral gabapentanoids, and once again, the heavy use of NSAIDs and opioid medications [9][10][11].
The main downfall of both approaches is the heavy reliance on opioids, a strong and addictive class of painkillers associated with massive negative side effects [12]. Opioids present a high risk for misuse and abuse, with rates of misuse having been shown as up to 56.3% and addiction rates as high as 23.0% in prescribed users [12]. Rates of opioid-dependence disorders, non-medical opioid use, unauthorized distribution, and accidental overdoses have all skyrocketed in the United States over the past 25 years [13].
Additionally, the United States is in the middle of the world's largest opioid epidemic [13][14][15][16][17][18], single-handedly consuming 80% of the world's opioid supply [20]. More than 4% of the countries' adult population currently misuses opioids [14], with 92% of these individuals misusing only physicianprescribed opioids and no other controlled substances [15], leading to an average of 130 Americans dying every day from opioid overdose [16]. Further, treatment of dependence, abuse, and overdose cases costs the United States economy up to $78.5 billion every year. Orthopedic surgery alone accounts for approximately 8.8% of postoperative opioid dependence cases in previously opioid-naive patients [19], prescribing more opioids than almost any other medical specialty [20].
The current multimodal analgesic approach mentioned above for both TKA and ACL reconstruction has attempted to combat this epidemic by lessening opioid use [7][8][9][10][11]; however, NSAIDs are the main alternative in these pain management strategies, which are also subject to severe negative consequences [21][22][23]. Harirforoosh et al. showed that NSAIDs could cause adverse effects in the gastrointestinal, cardiovascular, and renal systems [22] that limit or contraindicate NSAID use. NSAIDs have also been linked to postoperative bone and musculoskeletal soft tissue healing deficiencies, which is very concerning in orthopedic patients [21,23].
The combination of the opioid epidemic and lack of viable alternative, long-term pharmacologic analgesics highlights the demanding need for a non-pharmacologic analgesic intervention in the perioperative period. An alternative non-pharmacologic analgesic working to the same efficacy as opioids and NSAIDs could, in theory, completely replace said drugs to manage pain in TKA and ACL reconstruction operations, ending orthopedic surgeries' large contribution to the United States opioid crisis.
All of that said, non-pharmacologic perioperative analgesia is a very novel field of research, and the current literature is very limited. This systematic review intended to highlight and synthesize the available literature in the field, aiming to serve as a template for future research, effectively speeding up the process of establishing an efficacious non-pharmacologic analgesic for major knee surgery.
In doing so, the effectiveness of percutaneous peripheral nerve stimulation, percutaneous cyroneurolysis (percutaneous freezing of sensory nerves), and auricular acupressure were comparatively evaluated.

Systematic review methodology
A systematic search of the published literature was performed based on PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for studies published before July 9th, 2020.
A Medical Subject Heading (MeSH) plus keyword search was performed across PubMed, MEDLINE, Embase, Cochrane, and Web of Science. The MEDLINE search was conducted as follows: ((Exp Electric Stimulation Therapy/ OR Exp Audioanalgesia/ OR (Nonpharmacologic* OR "nonpharmacologic*" OR Cryoanalges* OR cryoneuro* OR Neuromodulat* OR electrostimulat* OR "electric* stimulation" OR "electrical muscle stimulation" OR "interferential current" OR electroacupuncture OR "nerve stimulation" OR audioanalges* OR analgesi* OR (pain adj2 manag*)).ti,ab) AND ((Exp Arthroplasty, Replacement, Knee/ OR ((exp Knee/ OR exp Knee Joint/) AND (exp Arthroplasty/)) OR TKA.ti,ab OR (knee adj2 (replacement* OR arthroplast*)).ti,ab) OR ((Exp Anterior Cruciate Ligament Injuries/ OR exp anterior cruciate ligament/ OR ACL.ti,ab OR "anterior cruciate ligament".ti,ab) AND (surgery.ti,ab OR reconstruction.ti, ab OR surgery.fs OR exp Surgical Procedures, Operative/)))) AND (perioperative.ti,ab OR "peri operative".ti,ab OR (immediate* adj2 (postoperative OR "post operative")).ti,ab OR exp Perioperative Care/ OR exp Perioperative Period/) The articles obtained were then evaluated based on strict inclusion and exclusion criteria. Studies were included that focused on (1) analyzing the efficacy of a non-pharmacologic peri-or immediate postoperative intervention for TKA or ACL reconstruction operations, (2) analyzing the effect of said interventions on short and long-term postoperative opioid use, (3) analyzing the clinical and functional outcomes of procedures in patients treated with said interventions, and (4) used human patients.
Studies were excluded based on the following criteria: (1) focusing on operations other than TKA or ACL reconstruction; (2) focusing on patients who were previously diagnosed with opioid addiction; (3) analyzing the efficacy of a pharmacologic (drug-based) intervention; (4) analyzing the efficacy of a nonpharmacologic intervention used in the rehabilitation phase (after the immediate postoperative period).
All studies were screened independently based on the previously stated criteria. Articles were screened initially by title and then abstract, after which the full text of all remaining manuscripts was reviewed. All full-text manuscripts determined as meeting criteria were included for analysis. Both groups showed downward trend in pain scores over the initial 5 min of stimulation when tested every minute and a similar increase in pain scores over the 5minute sham stimulation period. A mean pain score decrease of 84% was seen in both groups after 5 min of the secondary 30-minute stimulation period. Median pain score with movement was <2/10 by POD 4, and median pain score at rest across all patients was <4/10 on POD 1 and <2/10 by POD 3

-
No motor or sensory deficits reported in any patient Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram [42] illustrating the manuscript identification and screening process. The study was sham-controlled, and there is no mention of participants having knowledge of their intervention group. However, the treatment was selfadministered, so it is impossible to say with certainty that all participants were unaware they were performing a sham technique.

PN
The author specifically stated that participants were masked to their intervention throughout the trial. The surgeons and experimenters were unaware of participants' assigned interventions, but an acupuncturist had to train each participant to self-administer acupressure to either real or sham auricular acupressure points. The acupuncturist had no contact with the participants after training, and randomization was complete.
N All experimenters and providers were masked to the participants intervention during the entirety of the trial.   Domain 5: Risk of bias in selection of the reported results 5.1 Were the data that produced this result analyzed in accordance with a pre-specified analysis plan that was finalized before unblinded outcome data were available for analysis?
Number of participants needed for sufficient statistical power, dropout rates, and measures were discussed as if they were predetermined in the manuscript, but no protocol found.

PY
No statistical analysis was run because of the small sample size associated with it being a feasibility study. All listed measures in the methods are reported on and all data was analyzed before the researchers were unblinded to participants intervention type. The rating remains PN, however, as no statistical plan is listed or cited outside of the article itself.

PN
As no statistical analysis was run, it would not have been possible for the author to leave out certain representations of the statistical results.
N Risk-of-bias judgement: Domain 5 Low risk Rated as "some concern" because of NI response to 5.1.

Some Concern
Overall Risk of Bias Risk-of-bias judgement: All Low risk All but 1 domain is rated as low risk, but in accordance with the Cochrane risk-of-bias tool, the overall risk-of-bias is some concern.
Some Concern  Patients undergoing TKA were offered novel analgesic treatment and those that consented to the treatment were included in the report.

Yes
The included participants fit inclusion criteria that wellrepresents the population of those who require TKA  While no specific parts of the study design imply any greater risk-of-bias than all case series, the study is rated as "some concern" because no information was provided on dropouts or detailed enrollment parameters

Some Concern
This study was rated as some concern because the study design and inclusion/exclusion criteria were clear and pre-registered, but no information was provided on consecutive/inconsecutive participant enrollment, and no control group was used Some Concern While no specific parts of the study design imply any greater risk-of-bias than all case series, the study is rated as "some concern" because no information was provided on dropouts or detailed enrollment parameters Some Concern *Possible Responses: Yes, No, N/A, unclear.

Outcomes of interest
The primary outcome of interest in the presented study was patients' postoperative opioid use. Secondary outcomes of interest includes non-opioid postoperative medication use, pain scores, and clinical and functional surgical outcomes. These measures were intended to determine the ramifications of non-pharmacologic analgesia usage on the effectiveness of each operation.

Quality of evidence
Each of the six included studies was also assessed for riskof-bias: randomized controlled trials (RCTs) were assessed using the Revised Cochrane Risk-of-Bias Tool for Randomized Trials [24], case series were assessed using the Checklist for Quality Assessment of Non-Randomized Studies developed by Brazzelli et al. [25], and the retrospective chart review was assessed using Cochranes Tool to Assess Risk-of-Bias in Cohort Studies [26]. This assessment breakdown is represented visually in Tables 2-4. Overall potential risk-of-bias ratings for each study included low risk, high risk, or some concern [24].
Subsequently, the overall body of evidence was assessed using the GRADE (Grades of Recommendation, Assessment, Development, and Evaluation) criteria for each primary and secondary outcome. As such, the authors evaluated the evidence for each outcome based on study type, risk-of-bias, and the precision, consistency, and directness of the data. All GRADE assessment categories and outcomes evaluated can be seen in Table 5. The potential GRADE score for each outcome ranged from high quality to very low quality. Further, each outcome was rated as critical, important, or not important.

Study selection
Database searches resulted in a total of 1902 publications to be screened. Initially, duplicates were removed, resulting in 1354 remaining papers. These papers were first screened by titlewhere 1116 papers were excluded, then abstractwhere 211 more papers were excluded. Finally, full-text manuscripts of the remaining 27 publications were screened, and 6 were deemed to fit inclusion criteria. Figure 1 illustrates the screening process.

Quality of evidence
The six included studies consisted of two RCTs [27,28], three case series [29][30][31], and one retrospective chart review [32]. The risk-of-bias in each study were all determined to be either "low risk" or "some concern" (Tables 2-4).
GRADE quality of evidence was assessed for five separate outcomes: postoperative opioid use, other postoperative analgesic use, pain scores, clinical outcomes, and functional outcomes. The quality of evidence was rated as "low" for all, primarily due to the small sample sizes of most included studies, and all outcomes were graded as either "critical" or "important" (Table 5). Dasa et al. [32] Signaling question Comments Response Was selection of exposed and non-exposed cohorts drawn from the same population?
The exposed and unexposed cohorts were drawn from the same site and were all consecutive patients; however, the 50 unexposed participants all received treatment directly before the 50 exposed participants PY Can we be confident in the assessment of exposure?
Cohorts were drawn from directly before and after the implementation of a new intervention protocol at the study site DY Can we be confident that the outcome of interest was not present at the start of the study?
The outcomes of interest are all postoperative measures DY Did the study match exposed and unexposed for all variables that are associated with the outcome of interest or did the statistical analysis adjust for these prognostic variables?
Yes, all variables associated with the outcome of interest were analyzed, and no statistically significant differences were found  2 randomized controlled trials, 3 case series; 1 retrospective chart review Some ROB concern for 1 RCT because no statistical analysis was run due to the small sample size; some ROB concern for the case series due to lacking information regarding participant enrollment and indirect control groups; no serious concerns for chart review

No serious inconsistencies
Some concern because the populations studied differed: RCTs vs. case series vs. retrospective chart review; measure was comparable across study Small sample size (n) in 4/6 studies

None Low Critical
Clinical outcomes 4 1 randomized controlled trial, 2 case series, 1 retrospective chart review No serious ROB concern for RCT; some ROB concern for the case series due to lacking information regarding participant enrollment and indirect control groups; no serious ROB concern for chart review

No serious inconsistencies
Some concern because the populations studied differed: RCTs vs. case series vs. retrospective chart review; measure was comparable across study Small sample size (n) in 2/4 studies None Low Critical Functional Outcomes 5 2 randomized controlled trials, 3 case series; 1 retrospective chart review Some ROB concern for 1 RCT because no statistical analysis was run due to the small sample size; some ROB concern for the case series due to lacking information regarding participant enrollment and indirect control groups

No serious inconsistencies
Some concern because the populations studied differed: RCTs vs. case series; measure was comparable across study

Introduction to included studies: operations and non-pharmacologic analgesics investigated
Five included studies investigated the effects of non-pharmacologic analgesics on TKA patients [27,[29][30][31][32], while one studied ACL reconstruction patients [28]. Sample sizes ranged from a case series of 3 patients [31], to a 100-patient clinical trial [32]; details are shown in Table 1.
Three of these studies investigated the efficacy of percutaneous peripheral nerve stimulation, two percutaneous cryoneurolysis, and one auricular acupressure (Table 1). Lastly, the period of treatment and measurement ranged from five days preoperatively to six weeks postoperatively.

Postoperative use of opioids and other pharmacologic analgesics
As seen in Table 1, five of the six included studies showed a significant decrease in opioid use for the treatment group, as compared to controls [27,[29][30][31][32]; the final study did not directly measure opioid consumption compared to control groups, so statistical significance could not be obtained. Even so, only 50% of patients in the final study undergoing peripheral nerve stimulation requested simultaneous opioids [28].
All three non-pharmacologic interventions proved successful at reducing postoperative opioid consumption. That said, the most overall success was seen in the Ilfeld et al. percutaneous peripheral nerve stimulation study, in which they showed a median time to opioid cessation of only six days [29].
Only Ilfeld et al. in 2018 reported on other pharmacologic analgesic medication usage, showing that 7/10 patients elected to turn on a canal nerve block to supplement non-pharmacologic treatment. However, the same paper also showed more encouraging data in that 8/10 patients elected to use an optional perineural injection analgesic on postoperative days (POD) 1 and 2, but only 3/10 patients still required the same treatment on POD 3 [28], suggesting peripheral nerve stimulation quickly became an independently successful analgesic intervention.

Postoperative pain scores
All six included studies showed their respective interventions to be successful analgesics as measured by postoperative pain scores [27][28][29][30][31][32]. Only He et al. and Dasa et al. compared non-pharmacologic interventions to positive control groups, both reporting significantly lower pain scores for patients in the non-pharmacologic intervention condition [27,32]. That said, Dasa et al. showed that cryoneurolysis treatment significantly decreased pain scores at all postoperative time points measured. Comparatively, He et al. auricular acupressure treatment was no more successful than the control for the first 48 h after surgery but reached statistical significance from day 3 onward, suggesting cryoneurolysis may be superior to auricular acupressure in the direct perioperative period [27,32]. All other included studies reported mild pain scores for almost all treatment group patients (defined as <4/10 on a self-report pain scale) [28][29][30][31], the details of these results are outlined in Table 1.

Clinical and functional outcomes
Clinical outcomes should be considered very important secondary outcomes for novel non-pharmacologic perioperative analgesics because it ensures that any successes with respect to reduced opioid consumption and pain management do not come at the expense of the primary operational goal. Four included studies reported data on the clinical outcome of each operation [27,29,30,32]. As seen in Table 1, all four studies showed significant clinical improvement after surgery [27,29,30,32]. While the variability in reporting mechanism makes direct comparisons between non-pharmacologic analgesic interventions difficult, consistency over time suggests that cryoneurolysis and percutaneous peripheral nerve stimulation are more successful than auricular acupressure in promoting positive clinical outcomes [27,29,30,32], albeit to a small degree. Notably, none of the interventions negatively affect clinical outcomes [27,29,30,32].

Discussion
Non-pharmacologic analgesic interventions reduce opioid use and manage peri-and postoperative pain Cryoneurolysis, percutaneous peripheral nerve stimulation, and auricular acupressure all have the capability to successfully reduce postoperative opioid use in major knee surgery [27][28][29][30][31][32]. While small sample sizes make a direct comparison difficult, it appears that peripheral nerve stimulation may have a small primary advantage in reducing opioid consumption over the other two interventions. This interpretation is primarily based on the data (Table 1) showing that percutaneous peripheral nerve stimulation leads to a slim majority of patients (13/25 total) discontinuing opioid use within the first postoperative week or not requesting opioids at all [28][29][30]. The stated timeline is more drastic than any time to opioid cessation data from cryoneurolysis or auricular acupressure (though cryoneurolysis follows closely behind). However, the effect is almost certainly not significant, and there is a degree of variation within percutaneous peripheral nerve stimulation studiesone having shown a median time to opioid cessation of 6 days [29], and another 16.5 days [30]. Previous large-scale RCTs have shown the control group median time to opioid cessation in orthopedic surgery to range from 32 to 39 days postoperatively [33,34], suggesting that the presented studies show large improvement from baseline levels. Moreover, it makes many patients who stopped taking opioids within the first week or forewent opioids completely more noteworthy.
Time to cessation of opioid use is an extremely important measure of non-pharmacologic analgesic success because opioid use of more than 31 days postoperatively is associated with a greater likelihood of chronic opioid use one year after surgery [35]. All non-pharmacologic analgesics assessed in the present study proved to be opioid-sparing [27][28][29][30][31][32]. Therefore, this data does not necessarily exhibit any superiority of results amongst the three modalities reviewed in this paper, but heavily emphasizes the advantages of utilizing any of these interventions over opioids.
Similarly, cryoneurolysis and percutaneous peripheral nerve stimulation reduced postoperative pain scores earlier in the perioperative period than did auricular acupressure [27][28][29][30][31][32]. Auricular acupressure reduced pain scores at POD 3 and beyond [27], but cryoneurolysis and percutaneous peripheral nerve stimulation did so immediately on POD 1 [28][29][30][31][32]. This early pain relief may have greater implications than just patient comfort, as higher pain on the day of surgery is a significant predictor of chronic, postoperative opioid use [36]. Thus, an even greater opioid reduction advantage may be seen in implementing cryoneurolysis or percutaneous peripheral nerve stimulation over auricular acupressure.

Surgical outcomes
It is inherently important to ensure any novel analgesic does not negatively affect surgical outcomes. As such, it is of great significance that none of the presented non-pharmacologic interventions worsened clinical or functional outcomes following major knee surgery [27][28][29][30][31][32].
One major difference is noteworthy in that cryoneurolysis was the only modality to show improved clinical outcomes over a control group [32]; with significant improvements seen in the treatment group for KOOS Symptoms Subscale scores and postoperative length of stay [32]. Alternatively, percutaneous peripheral nerve stimulation and auricular acupressure only showed significant improvement in clinical or functional outcomes compared to preoperative baselines [27][28][29][30]. While such improvement certainly maintains the analgesic viability of percutaneous peripheral nerve stimulation and auricular acupressure, it simultaneously indicates a distinct advantage to cryoneurolysis. The identified gap in clinical outcomes can most likely be attributed to the robust ability of cryoneurolysis to decrease pain in the perioperative period [31,32], avoiding uncontrolled pain, which leads to slower patient mobilization and a delayed start to rehabilitation, worsening clinical outcomes after TKA and ACL reconstruction [37,38]. It has also been shown that preemptive analgesia can lessen the postoperative inflammatory response and perioperative pain [39,40]. Fittingly, cryoneurolysis is the only of the three non-pharmacologic analgesics analyzed to be fully administered prior to surgery [29,31].

The most efficacious non-pharmacologic analgesic interventions
As alluded to above, cryoneurolysis, percutaneous peripheral nerve stimulation, and auricular acupressure appear to all be viable analgesic interventions for TKA and ACL reconstruction surgeries. That said, the presented data suggest that cryoneurolysis, closely followed by percutaneous peripheral nerve stimulation, maybe the most beneficial to implement in clinical settings.
Cryoneurolysis is the most successful intervention at inhibiting peri-and postoperative pain [31,32]. Both cryoneu-rolysis and percutaneous peripheral nerve stimulation showed immediate postoperative analgesic efficacy [28][29][30][31][32], but cryoneurolysis separates as the most efficacious intervention because patients in such trials also exhibited slightly lower pain scores than those who underwent percutaneous peripheral nerve stimulation, while maintaining significant decreases in postoperative opioid use [31,32]. Cryoneurolysis is also the only analyzed non-pharmacologic analgesic to improve clinical outcomes [32]. As detailed in Tables 2-5, cryoneurolysis also shows the strongest quality of evidence for included interventions. Combined with the factors described above, this leads to the conclusion that cryoneurolysis, followed in order by percutaneous peripheral nerve stimulation and auricular acupressure, is the most promising novel perioperative nonpharmacologic analgesic intervention for implementation into clinical use.

Limitations of the presented studies
While considering all the presented data, it is also important to recognize the limitations of each study. Most limitations in the presented systematic review stem from the novelty of perioperative non-pharmacologic analgesic research. The largest overall limitation is the small number of suitable studies and small sample sizes investigated in said studies. As shown in Table 1, the two included studies were relatively large, with over 90 participants, but the remainder all had 18 or fewer. The small sizes led to other limitations, such as one study not having attempted to determine statistical significance [28], and two others having elected to statistically compare novel interventions to historical controls instead of a positive control group [30,31]. Lastly, four of the six included studies were performed by the same first author [28][29][30][31]. While that fact limits the generalizability of the presented conclusions, the authors still deemed it apt to include all four of the studies for two reasons: (1) given the novelty of the field and a small amount of available literature, these studies also make up much of the literature regarding perioperative non-pharmacologic analgesics, and (2) all four of the studies met the systematic review's inclusion criteria.

Implications of the data
As highlighted in the Introduction section, the primary benefit of a successful perioperative non-pharmacologic analgesic would be to replace opioids as the primary postoperative analgesic in major knee surgery. Doing so would eliminate the vast negative effects of opioid use, including concerningly high misuse and abuse rates in prescribed users and a high frequency of adverse side effects [12]. Moreover, this is of particular importance to orthopedic researchers as the United States is currently in the middle of the world's worst opioid epidemic [13][14][15][16][17][18], with opioid addiction and overdose rates rising rapidly over the last quarter-century and orthopedic surgery being the single largest prescribing medical specialty [19,20]. Fortunately, the data presented in this systematic review indicates that several non-pharmacologic interventions may be viable to replace opioids as the primary perioperative analgesic for TKA and ACL reconstruction operations.

Direction of future research
Future research into perioperative non-pharmacologic analgesics is certainly warranted. The first and most telling research that should be performed is to implement full-scale, multi-site RCTs for cryoneurolysis and percutaneous peripheral nerve stimulation. While the data from pilot studies are very promising, RCTs would drastically improve the reliability of results seen and overcome the available literature's current issue with small sample sizes. Moreover, it would be very telling to measure longer-term follow-ups, allowing researchers to determine if there are any negative long-term effects on nerve function from either cryoneurolysis or percutaneous peripheral nerve stimulation. Although studies utilizing cryoneurolysis to treat chronic pain have shown it to be safe and reversible over time [41], so completing these long-term studies may not be as pressing.
Other interesting extensions of the presented studies would be determining if an additive or synergistic effect can be seen by simultaneously implementing auricular acupressure and either cryoneurolysis or percutaneous peripheral nerve stimulation. Lastly, a notable distinction exists in the presented papers evaluating cryoneurolysis and percutaneous peripheral nerve stimulation that some utilize ultrasound guidance for implementation, and some do not. While it seems reasonable that ultrasound guidance is safer and allows for more detailed lead placement, similar results were seen in the presented review for studies including or excluding ultrasound guidance. Hence, a trial directly comparing outcomes for these groups would be important in setting clinical best practices.

Conclusion
The present systematic review aimed to analyze and compare the available literature on perioperative non-pharmacologic analgesic interventions in the hopes of advancing a novel field of orthopedic research, with the end goal of lessening postoperative opioid use after TKA and ACL reconstruction operations. As such, the authors conclude that cryoneurolysis, percutaneous peripheral nerve stimulation, and auricular acupressure all appear to be viable and successful non-pharmacologic analgesics in the perioperative period. Of the three, cryoneurolysis seems the most promising moving into future research.